Introduction to Embedding Space

Sentence Examples

Discover more insights into Embedding Space

Keywords frequently search together with Embedding Space

Narrow sentence examples with built-in keyword filters

Embedding Space sentence examples within convolutional neural network

convolutional neural network

The distances are computed in a feature embedding space where the scans are mapped by a convolutional neural network (CNN).

Loop-closure detection by LiDAR scan re-identification

convolutional neural network

A multi-stream convolutional neural network (CNN) framework is proposed to effectively learn the 3D volume and 2D MIP feature vectors, respectively, and then explore their inter-dependencies in a joint volume-composition embedding space by unprojecting the 2D feature vectors into the 3D volume embedding space.

VC-Net: Deep Volume-Composition Networks for Segmentation and Visualization of Highly Sparse and Noisy Image Data

More convolutional neural network sentences More Sentences

Embedding Space sentence examples within cross modal retrieval

cross modal retrieval

Then, armed with this dataset, we describe several approaches which leverage scene text, including a better scene-text aware cross-modal retrieval method which uses specialized representations for text from the captions and text from the visual scene, and reconcile them in a common embedding space.

StacMR: Scene-Text Aware Cross-Modal Retrieval

cross modal retrieval

In this paper, we propose a combination of both problems into a continual cross-modal retrieval setting, where we study how the catastrophic interference caused by new tasks impacts the embedding spaces and their cross-modal alignment required for effective retrieval.

Continual learning in cross-modal retrieval

More cross modal retrieval sentences More Sentences

Embedding Space sentence examples within pre trained word

pre trained word

Specifically, we learn an encoder to generate a debiased version of an input word embedding such that it (a) retains the semantics of the pre-trained word embedding, (b) agrees with the unbiased definition of the word according to the dictionary, and (c) remains orthogonal to the vector space spanned by any biased basis vectors in the pre-trained word embedding space.

Dictionary-based Debiasing of Pre-trained Word Embeddings

pre trained word

The novel contributions of our solution include: (i)the introduction of a novel data representation for topic modeling based on syntactic and semantic relationships derived from distances calculated within a pre-trained word embedding space and (ii)the proposal of a new TF-IDF-based strategy, particularly developed to weight the CluWords.

CluWords: Exploiting Semantic Word Clustering Representation for Enhanced Topic Modeling

More pre trained word sentences More Sentences

Embedding Space sentence examples within Joint Embedding Space

Joint Embedding Space

Our network consists of two components: a generator to synthesize gestures from a joint embedding space of features encoded from the input speech and the seed poses, and a discriminator to distinguish between the synthesized pose sequences and real 3D pose sequences.

Speech2AffectiveGestures: Synthesizing Co-Speech Gestures with Generative Adversarial Affective Expression Learning

Joint Embedding Space

Towards overcoming this challenge, we propose Hierarchical Moment Alignment Network (HMAN) which learns an effective joint embedding space for moments and sentences.

Text-based Localization of Moments in a Video Corpus

More Joint Embedding Space sentences More Sentences

Embedding Space sentence examples within Dimensional Embedding Space

Dimensional Embedding Space

Most existing methods remove batch effects in a low-dimensional embedding space.

A joint deep learning model enables simultaneous batch effect correction, denoising and clustering in single-cell transcriptomics.

Dimensional Embedding Space

The series of adverse weather events are embedded into a high-dimensional embedding space to enable the identification of temporal event patterns similar to those observed in risky areas.

Estimating the incidence of adverse weather effects on road traffic safety using time series embeddings

More Dimensional Embedding Space sentences More Sentences

Embedding Space sentence examples within Word Embedding Space

Word Embedding Space

Our hypothesis is that a word's representations in both word embedding spaces are more similar for non-biased words than biased words.

Identification of Biased Terms in News Articles by Comparison of Outlet-Specific Word Embeddings

Word Embedding Space

We apply an established retrofitting method to harness the verb class membership knowledge from BioVerbNet and transform a pretrained word embedding space by pulling together verbs belonging to the same semantic-syntactic class.

BioVerbNet: a large semantic-syntactic classification of verbs in biomedicine

More Word Embedding Space sentences More Sentences

Embedding Space sentence examples within Latent Embedding Space

Latent Embedding Space

The proposed DLCL module lends on the idea of latent concepts to learn compact representations in the latent embedding space in an unsupervised way.

Unsupervised Discriminative Embedding for Sub-Action Learning in Complex Activities

Latent Embedding Space

Despite successes in transferring the style based on the encoder-decoder framework, current approaches still lack the ability to preserve the content and even logic of original sentences, mainly due to the large unconstrained model space or too simplified assumptions on latent embedding space.

GTAE: Graph-Transformer based Auto-Encoders for Linguistic-Constrained Text Style Transfer

More Latent Embedding Space sentences More Sentences

Embedding Space sentence examples within Common Embedding Space

Common Embedding Space

Here, for this task we propose using metric learning, where a common embedding space for sessions and items is created, and distance measures dissimilarity between the provided sequence of users’ events and the next action.

Metric Learning for Session-based Recommendations

Common Embedding Space

Then, armed with this dataset, we describe several approaches which leverage scene text, including a better scene-text aware cross-modal retrieval method which uses specialized representations for text from the captions and text from the visual scene, and reconcile them in a common embedding space.

StacMR: Scene-Text Aware Cross-Modal Retrieval

More Common Embedding Space sentences More Sentences

Embedding Space sentence examples within Semantic Embedding Space

Semantic Embedding Space

We present a method for defining warmth and competence axes in semantic embedding space, and show that the four quadrants defined by this subspace accurately represent the warmth and competence concepts, according to annotated lexicons.

Understanding and Countering Stereotypes: A Computational Approach to the Stereotype Content Model

Semantic Embedding Space

To ensure the model's extendibility, it embeds candidate answers and recognized texts in a semantic embedding space and adopts semantic embedding based classifier to perform answer prediction.

Dual Path Multi-Modal High-Order Features for Textual Content based Visual Question Answering

More Semantic Embedding Space sentences More Sentences

Embedding Space sentence examples within Learned Embedding Space

Learned Embedding Space

Then, in the domain adaptation process, beyond domain alignment, we employ Laplacian Eigenmaps to ensure the domain structure is consistently distributed in the learned embedding space.

Domain structure-based transfer learning for cross-domain word representation

Learned Embedding Space

It formalizes an end-to-end network architecture, referred to as b-Net, which accomplishes noise suppression through attention masking in a learned embedding space.

Speech Enhancement via Attention Masking Network (SEAMNET): An End-to-End System for Joint Suppression of Noise and Reverberation

More Learned Embedding Space sentences More Sentences

Embedding Space sentence examples within Feature Embedding Space

Feature Embedding Space

The distances are computed in a feature embedding space where the scans are mapped by a convolutional neural network (CNN).

Loop-closure detection by LiDAR scan re-identification

Feature Embedding Space

Among different few-shot learning algorithms, metric-based methods, which focus on how to acquire a robust feature embedding space or distance metric for classification, are most studied and believed to be effective for classification tasks.

Task-adaptive Relation Dependent Network for Few-shot Learning

More Feature Embedding Space sentences More Sentences

Embedding Space sentence examples within Shared Embedding Space

Shared Embedding Space

Visual-semantic embedding (VSE) networks create joint image-text representations to map images and texts in a shared embedding space to enable various information retrieval-related tasks, such as image-text retrieval, image captioning, and visual question answering.

On the Limitations of Visual-Semantic Embedding Networks for Image-to-Text Information Retrieval

Shared Embedding Space

More specifically, our work o↵ers (i) an approach to learn neural representations of product attribute-values within a shared embedding space as product reviews; (ii) a weighted composition strategy to develop product representations from the representation of its attributes; and, (iii) a review selection method that selects relevant reviews for the composed product representation within the neural embedding space.

Learning product representations for generating reviews for cold products

More Shared Embedding Space sentences More Sentences

Embedding Space sentence examples within Unified Embedding Space

Unified Embedding Space

Existing methods usually learn a similarity mapping of local parts between image and text, or embed the whole image and text into a unified embedding space.

Divide-and-Merge the embedding space for cross-modality person search

Unified Embedding Space

The semantic-based approach maps mentions and entities in different languages to a unified embedding space, which reduces dependence on large-scale bilingual dictionaries.

Pivot-based Candidate Retrieval for Cross-lingual Entity Linking

More Unified Embedding Space sentences More Sentences

Embedding Space sentence examples within Discriminative Embedding Space

Discriminative Embedding Space

To this end, we first learn a discriminative embedding space for vehicle-terrain interaction sounds from triplets of audio clips formed using visual features of the corresponding terrain patches and cluster the resulting embeddings.

Self-Supervised Visual Terrain Classification From Unsupervised Acoustic Feature Learning

Discriminative Embedding Space

In this work, we propose to seek an angularly discriminative Embedding Space where representations are continuous and have smaller intra-class angular margin and larger inter-class angular margin.

Learning to Hash based on Angularly Discriminative Embedding

More Discriminative Embedding Space sentences More Sentences

Embedding Space sentence examples within Specific Embedding Space

Specific Embedding Space

Based on these modules, we propose a hierarchical attribute-aware embedding network (HAEN) which takes images and attributes as input, learns multiple attribute-specific embedding spaces, and measures fine-grained similarity in the corresponding spaces.

Learning Fashion Similarity Based on Hierarchical Attribute Embedding

Specific Embedding Space

Fashion retrieval methods aim at learning a clothing-specific embedding space where images are ranked based on their global visual similarity with a given query.

Localized Triplet Loss for Fine-grained Fashion Image Retrieval

More Specific Embedding Space sentences More Sentences

Embedding Space sentence examples within Monolingual Embedding Space

Monolingual Embedding Space

These GANs-based methods enable the alignment of two monolingual embedding spaces approximately, but the performance on the embeddings of low-frequency words (LFEs) is still unsatisfactory.

Adversarial training with Wasserstein distance for learning cross-lingual word embeddings

Monolingual Embedding Space

One mainstream method in cross-lingual word embeddings is to learn a linear mapping between two monolingual embedding spaces using a training dictionary.

Learning Cross-Lingual Mappings in Imperfectly Isomorphic Embedding Spaces

More Monolingual Embedding Space sentences More Sentences

Embedding Space sentence examples within Deep Embedding Space

Deep Embedding Space

We present Motion2Vec that learns a deep embedding space by minimizing a metric learning loss in a Siamese network: images from the same action segment are pulled together while being pushed away from randomly sampled images of other segments, and a time contrastive loss is used to preserve the temporal ordering of the images.

Sequential robot imitation learning from observations

Deep Embedding Space

Metric-learning-based methods, which attempt to learn a deep embedding space on extremely large episodes, have been successfully applied to few-shot classification problems.

Enhancing Metric-Based Few-Shot Classification With Weighted Large Margin Nearest Center Loss

More Deep Embedding Space sentences More Sentences

Embedding Space sentence examples within Multimodal Embedding Space

Multimodal Embedding Space

, text-to-image) by jointly learning a multimodal embedding space.

Multimodal Metric Learning for Tag-Based Music Retrieval

Multimodal Embedding Space

We then integrate coarse‐grained and fine‐grained features into a multimodal embedding space, in which the similarity degrees between images and texts can be directly compared.

Cross‐modal retrieval with dual multi‐angle self‐attention

More Multimodal Embedding Space sentences More Sentences

Embedding Space sentence examples within Invariant Embedding Space

Invariant Embedding Space

Specifically, we optimize the differences between the embeddings of a support set and a query set in order to learn a channel-invariant embedding space for utterances.

Meta-Learning for Cross-Channel Speaker Verification

Invariant Embedding Space

We introduce in this letter a dual-stream mutual information distillation network (MIND-Net), whereby the non-identity specific mutual information (MI) characterized by generic face features coexistent on realistic and synthetic LR face images are distilled to render a resolution-invariant embedding space for LRFR.

MIND-Net: A Deep Mutual Information Distillation Network for Realistic Low-Resolution Face Recognition

More Invariant Embedding Space sentences More Sentences

Embedding Space sentence examples within Visual Embedding Space

Visual Embedding Space

This problem is especially pronounced in data-scarce settings where the data is relatively small (10% of the large scale MSR-VTT) to cover the rather complex audio-visual embedding space.

Cross Lingual Video and Text Retrieval: A New Benchmark Dataset and Algorithm

Visual Embedding Space

By minimizing the cross-entropy loss between a prediction and a given class label, the NN and its visual embedding space are learned to fulfill a given task.

Learning Visual Models Using a Knowledge Graph as a Trainer

More Visual Embedding Space sentences More Sentences

Embedding Space sentence examples within Cell Embedding Space

Cell Embedding Space

In this paper, we propose a deep neural network model which can embed semantic and contextual information about tabular cells in a low-dimensional cell embedding space.

Learning cell embeddings for understanding table layouts

Cell Embedding Space

In this paper, we propose a method to embed the semantic and contextual information about tabular cells in a low dimension cell embedding space.

Tabular Cell Classification Using Pre-Trained Cell Embeddings

More Cell Embedding Space sentences More Sentences

Embedding Space sentence examples within Contextual Embedding Space

Contextual Embedding Space

synonyms, paraphrases), allowing translations that are different from the references, yet close in the contextual embedding space, to be treated as substantially correct.

BERTTune: Fine-Tuning Neural Machine Translation with BERTScore

Contextual Embedding Space

We propose a novel method to estimate polysemy based on simple geometry in the contextual embedding space.

Unsupervised Word Polysemy Quantification with Multiresolution Grids of Contextual Embeddings

More Contextual Embedding Space sentences More Sentences

Embedding Space sentence examples within Spherical Embedding Space

Spherical Embedding Space

Besides, we use a generative directional appearance module to estimate and dynamically update the foreground/background class probabilities in a spherical embedding space.

4G-VOS: Video Object Segmentation using guided context embedding

Spherical Embedding Space

Second, an effective directional appearance model-based statistics is proposed to represent the target and background on a spherical embedding space for VOS.

Directional Deep Embedding and Appearance Learning for Fast Video Object Segmentation

More Spherical Embedding Space sentences More Sentences

Embedding Space sentence examples within Continuou Embedding Space

Continuou Embedding Space

We have developed a novel encoder-decoder network termed as DVICE (Directed Variational Inference Cross Encoder), which learns a continuous embedding space to ensure better similarity learning.

Directed Variational Cross-encoder Network for Few-shot Multi-image Co-segmentation

Continuou Embedding Space

First, we propose OptiPrompt, a novel and efficient method which directly optimizes in continuous embedding space.

Factual Probing Is [MASK]: Learning vs. Learning to Recall

More Continuou Embedding Space sentences More Sentences

Embedding Space sentence examples within Speaker Embedding Space

Speaker Embedding Space

Unlike the conventional back-end augmentation method which adds noises to the raw audios and then extracts augmented embeddings, in this work, we proposed a noise distribution matching (NDM) based algorithm in the speaker embedding space.

Speaker Embedding Augmentation with Noise Distribution Matching

Speaker Embedding Space

Lastly, we investigate if the learned multi-channel speaker embedding space can be made more discriminative through a contrastive loss-based fine-tuning.

Multi-Channel Speaker Verification for Single and Multi-Talker Speech

More Speaker Embedding Space sentences More Sentences

Embedding Space sentence examples within Sentence Embedding Space

Sentence Embedding Space

In this work, we propose an alternative approach to achieving video moment retrieval that requires no textual annotations of videos and instead leverages the existing visual concept detectors and a pre-trained image-sentence embedding space.

Learning Video Moment Retrieval Without a Single Annotated Video

Sentence Embedding Space

We explore augmentation in the sentence embedding space as well as prototypical embedding space.

Protoda: Efficient Transfer Learning for Few-Shot Intent Classification

More Sentence Embedding Space sentences More Sentences

Embedding Space sentence examples within Suitable Embedding Space

Suitable Embedding Space

We propose EmbedNet, which is trained using a triplet loss for learning a suitable embedding space where the embedding of the word image lies closer to the embedding of the corresponding text transcription.

Improving Word Recognition using Multiple Hypotheses and Deep Embeddings

Suitable Embedding Space

Moreover, it adds the reconstruction loss to the objective function combining the dimensionality reduction with clustering to find a more suitable embedding space for clustering.

ScCAEs: deep clustering of single-cell RNA-seq via convolutional autoencoder embedding and soft K-means.

More Suitable Embedding Space sentences More Sentences

Embedding Space sentence examples within Learnt Embedding Space

Learnt Embedding Space

Given a new RGB frame, MOLTR firstly applies a monocular 3D detector to localize objects of interest and extract their shape codes that represent the object shape in a learnt embedding space.

MOLTR: Multiple Object Localization, Tracking and Reconstruction From Monocular RGB Videos

Learnt Embedding Space

In addition, they ensure the learnt embedding space possesses the property of proximity preservation.

Behavior regularized prototypical networks for semi-supervised few-shot image classification

More Learnt Embedding Space sentences More Sentences

Embedding Space sentence examples within Large Embedding Space

Large Embedding Space

By making full use of strong-correlation between adjacent groups, GCC can compress not only consecutive several groups whose bits are valued 1 (or 0) but also a single group so that a large embedding space is provided.

High capacity reversible data hiding in encrypted images using SIBRW and GCC

Large Embedding Space

The learned models are able to successfully and robustly identify the underlying modes governing the system, even with a redundantly large embedding space.

Modal extraction of spatiotemporal atomization data using a deep convolutional Koopman network

More Large Embedding Space sentences More Sentences

Embedding Space sentence examples within Emotion Embedding Space

Emotion Embedding Space

Both of them tend to explore the underlying semantic emotion information but with a shared recognition network or with a shared emotion embedding space, respectively.

EmoBed: Strengthening Monomodal Emotion Recognition via Training with Crossmodal Emotion Embeddings

Emotion Embedding Space

Emotion embedding space learned from references is a straight-forward approach for emotion transfer in encoder-decoder structured emotional text to speech (TTS) systems.

Controllable Emotion Transfer For End-to-End Speech Synthesis

More Emotion Embedding Space sentences More Sentences

Embedding Space sentence examples within embedding space vium

embedding space vium

Binaural input waveforms are mapped to a joint embedding space via a learned encoder, and separate multiplicative masking mechanisms are included for noise suppression and speaker separation.

Speaker separation in realistic noise environments with applications to a cognitively-controlled hearing aid

embedding space vium

In this paper, we propose a novel framework that models both queries and items in shared latent embedding space via a dual sequence transformer structure, which captures customer’s potential interest from the prospect of reconciling the historical queries and corresponding customers interactions.

Dual Sequence Transformer for Query-based Interactive Recommendation

More embedding space vium sentences More Sentences

Embedding Space sentence examples within embedding space instead

embedding space instead

Specifically, we first define the adversarial perturbations in the embedding space instead of in the discrete graph domain to circumvent the challenge of generating discrete adversarial examples.

Adversarial training regularization for negative sampling based network embedding

embedding space instead

To address this issue, we propose a novel multi-view clustering method that is able to construct an essential similarity graph in a spectral embedding space instead of the original feature space.

Consensus Graph Learning for Multi-view Clustering

More embedding space instead sentences More Sentences

More Embedding Space sentence examples
10.1007/s11263-021-01486-4

Our approach utilizes the semantic and temporally structured sub-motion prior in the embedding space for motion capture and understanding in a datadriven multi-task manner.

SportsCap: Monocular 3D Human Motion Capture and Fine-grained Understanding in Challenging Sports Videos

More Embedding Space sentence examples
10.18653/v1/2021.acl-long.407

In this paper, we propose a Knowledge-Guided Frame Identification framework (KGFI) that integrates three types frame knowledge, including frame definitions, frame elements and frame-to-frame relations, to learn better frame representation, which guides the KGFI to jointly map target words and frames into the same embedding space and subsequently identify the best frame by calculating the dot-product similarity scores between the target word embedding and all of the frame embeddings.

A Knowledge-Guided Framework for Frame Identification

More Embedding Space sentence examples
10.1007/s10489-020-02110-7

Basically, two crucial steps are involved in CDP: The first step is to find the critical directions for each category in the embedding space, and the second step is to measure the similarity between samples and critical directions according to the projection length.

Critical direction projection networks for few-shot learning

More Embedding Space sentence examples
10.18653/v1/2021.acl-srw.35

This paper sheds some light on the embedding spaces they create, using average cosine similarity, contextuality metrics and measures for representational similarity for comparison, revealing that BERT and NMT encoder representations look significantly different from one another.

On the differences between BERT and MT encoder spaces and how to address them in translation tasks

More Embedding Space sentence examples
10.1016/J.POETIC.2021.101567

First, one can hold terms constant and measure how the embedding space moves around them—much like astronomers measured the changing of celestial bodies with the seasons.

Cultural cartography with word embeddings

More Embedding Space sentence examples
10.1007/jhep10(2021)141

We generalize the color/kinematics duality of flat-space scattering amplitudes to the embedding space formulation of AdS boundary correlators.

BCJ amplitude relations for Anti-de Sitter boundary correlators in embedding space

More Embedding Space sentence examples
10.18653/V1/2021.NAACL-MAIN.272

Given training instances consisting of input sentences and the descriptions of their seen relations, ZS-BERT learns two functions that project sentences and relations into an embedding space by jointly minimizing the distances between them and classifying seen relations.

ZS-BERT: Towards Zero-Shot Relation Extraction with Attribute Representation Learning

More Embedding Space sentence examples
10.1109/CVPRW53098.2021.00398

To this end, we contribute Supervised Contrastive Replay (SCR), which explicitly encourages samples from the same class to cluster tightly in embedding space while pushing those of different classes further apart during replay-based training.

Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learning

More Embedding Space sentence examples
10.1007/978-3-030-72113-8_30

In the second stage, we find the closest concept name representation in an embedding space to a given clinical mention.

Drug and Disease Interpretation Learning with Biomedical Entity Representation Transformer

More Embedding Space sentence examples
10.1101/2021.04.06.438536

Here, we present SCALEX, a deep-learning method that integrates single-cell data by projecting cells into a batch-invariant, common cell-embedding space in a truly online manner (i.

Online single-cell data integration through projecting heterogeneous datasets into a common cell-embedding space

More Embedding Space sentence examples
10.1109/TPAMI.2021.3113270

Deep metric learning aims at learning a mapping from the input domain to an embedding space, where semantically similar objects are located nearby and dissimilar objects far from another.

Improving Deep Metric Learning by Divide and Conquer

More Embedding Space sentence examples
10.1007/s11263-020-01421-z

However, due to the various appearances of rain streaks and accumulation, it is difficult to separate rain and background information in the embedding space, which results in inaccurate deraining.

Selective Wavelet Attention Learning for Single Image Deraining

More Embedding Space sentence examples
10.18653/v1/2021.eacl-main.184

Additionally, we present a method for creating lexical representations that capture connotations within the embedding space and show that using the embeddings provides a statistically significant improvement on the task of stance detection when data is limited.

A Unified Feature Representation for Lexical Connotations

More Embedding Space sentence examples
10.1145/3460231.3474233

Existing approaches focus on learning an embedding space where fashion items from different categories that are visually compatible are closer to each other.

Semi-Supervised Visual Representation Learning for Fashion Compatibility

More Embedding Space sentence examples
10.1109/ACCESS.2021.3096819

We train vocal and accompaniment encoders to learn a joint-embedding space of vocal and accompaniment tracks, where the embedded feature vectors of a compatible pair of vocal and accompaniment tracks lie close to each other and those of an incompatible pair lie far from each other.

Vocal-Accompaniment Compatibility Estimation Using Self-Supervised and Joint-Embedding Techniques

More Embedding Space sentence examples
10.3389/fimmu.2021.680687

Applying Immune2vec along with machine learning approaches to patient data exemplifies how distinct clinical conditions can be effectively stratified, indicating that the embedding space can be used for feature extraction and exploratory data analysis.

Immune2vec: Embedding B/T Cell Receptor Sequences in ℝN Using Natural Language Processing

More Embedding Space sentence examples
10.21428/594757db.9e67a9f0

This paper presents an investigation of topic modeling in embedding spaces performances in the context of depression assessment.

Topic Modeling in Embedding Spaces for Depression Assessment

More Embedding Space sentence examples
10.1016/J.INS.2021.04.095

The bridge nodes with small proximity and located in different communities, should be similar in embedding space since they have the same surrounding structure.

Lower order information preserved network embedding based on non-negative matrix decomposition

More Embedding Space sentence examples
10.28995/2075-7182-2021-20-1188-1196

For each corpus, we train word embeddings, and we obtain linear transformations between the embedding spaces.

[email protected] Measuring Semantic Difference as per-word Variation in Aligned Semantic Spaces

More Embedding Space sentence examples
10.18653/v1/2021.eacl-demos.11

Given an (i) embedding space (users can choose between the predefined spaces or upload their own) and (ii) a bias specification (users can choose between existing bias specifications or create their own), DebIE can (1) compute several measures of implicit and explicit bias and modify the embedding space by executing two (mutually composable) debiasing models.

DebIE: A Platform for Implicit and Explicit Debiasing of Word Embedding Spaces

More Embedding Space sentence examples
10.1109/ICRA48506.2021.9561078

Loop closure candidates are found using a nearest-neighbor lookup in the embedding space.

Spherical Multi-Modal Place Recognition for Heterogeneous Sensor Systems

More Embedding Space sentence examples
10.1038/s42256-021-00361-8

We use simulated and real data to demonstrate that INSCT generates an embedding space that accurately integrates cells across experiments, platforms and species.

Integration of millions of transcriptomes using batch-aware triplet neural networks

More Embedding Space sentence examples
10.1038/s41598-021-92973-8

The graph entropy measurement quantifies the information of the graphs and constructs the embedding space.

Entropy-based dynamic graph embedding for anomaly detection on multiple climate time series

More Embedding Space sentence examples
10.1109/tkde.2021.3093773

Second, the VAE based models are not expressive enough to capture the true preference since VAE often yields an embedding space of a single modal, while in real world, user-item interactions usually exhibit multi-modality on user preference distribution.

Dual Adversarial Variational Embedding for Robust Recommendation

More Embedding Space sentence examples
10.1109/ISBI48211.2021.9433978

However, a useful task in histopathology embedding is to train an embedding space regardless of the magnification level.

Magnification Generalization For Histopathology Image Embedding

More Embedding Space sentence examples
10.1145/3474624.3477067

Our approach organizes the software requirements into clusters and sub-clusters according to requirement similarities in the embedding space.

Hierarchical Cluster Labeling of Software Requirements using Contextual Word Embeddings

More Embedding Space sentence examples
10.1103/PhysRevE.103.012305

Popular methods employed in network embedding either rely on implicit approximations of the principle of proximity preservation or implement it by enforcing the geometry of the embedding space, thus hindering geometric properties that networks may spontaneously exhibit.

Model-free hidden geometry of complex networks.

More Embedding Space sentence examples
10.1109/ICPR48806.2021.9412197

Are multimodal signals useful to understand and anticipate human actions if acquired from the user viewpoint? This paper proposes to build an embedding space where inputs of different nature, but semantically correlated, are projected in a new representation space and properly exploited to anticipate the future user activity.

Anticipating Activity from Multimodal Signals

More Embedding Space sentence examples
10.1109/ICPR48806.2021.9412479

These systems consist on a feature extraction block (extracts feature maps from images) followed by a spatial dimensionality reduction block (generates compact image representations from the feature maps) and an embedding generation module (projects the image representation to the embedding space).

Generalized Local Attention Pooling for Deep Metric Learning

More Embedding Space sentence examples
10.3390/sym13040729

Following the mathematics allows me to define this granulated space-time as the embedding space of the skeleton of a computational spin network inside a quantum computer.

Emergent Space-Time in a Bubble Universe

More Embedding Space sentence examples
10.1109/LSP.2021.3076358

The key idea is to compact the bona fide speech representation and inject an angular margin to separate the spoofing attacks in the embedding space.

One-Class Learning Towards Synthetic Voice Spoofing Detection

More Embedding Space sentence examples
10.1145/3474085.3475347

Specifically, our approach consists of three ideas: (i) extracting compression-insensitive features from both uncompressed and compressed forgeries using an adversarial learning strategy; (ii) learning a robust partition by constructing a metric loss that can reduce the distance of the paired original and compressed images in the embedding space; (iii) improving the accuracy of tampered localization with an attention-transfer module.

Metric Learning for Anti-Compression Facial Forgery Detection

More Embedding Space sentence examples
10.1016/j.neucom.2020.09.065

While semantic vectors were mainly used to constrain the distance between different classes in embedding space in existing ZSL methods, we first emphasize the importance of the relationship between different dimensions in semantic space.

Disentangled features with direct sum decomposition for zero shot learning

More Embedding Space sentence examples
10.3390/ELECTRONICS10121372

This research presents a suite of data-efficient deep learning approaches to deduce the transfer function from the embedding space of English to that of Tamil, deploying three popular embedding algorithms: Word2Vec, GloVe and FastText.

Generation of Cross-Lingual Word Vectors for Low-Resourced Languages Using Deep Learning and Topological Metrics in a Data-Efficient Way

More Embedding Space sentence examples
10.3390/ijms22094632

Since the embedding distribution difference is reduced, embedding space analysis demonstrates that the model’s outputs of natural products are reliable.

Target Prediction Model for Natural Products Using Transfer Learning

More Embedding Space sentence examples
10.1007/978-3-030-86517-7_5

Contrastive learning can learn better representations by keeping the instances of each class closer and pushing away instances from different classes in the embedding space concurrently.

ConCAD: Contrastive Learning-based Cross Attention for Sleep Apnea Detection

Learn more from Embedding Space

Keywords related to Embedding

Embedding Space
Encyclopedia

What is/are Embedding Space?

Embedding Space - Our approach utilizes the semantic and temporally structured sub-motion prior in the embedding space for motion capture and understanding in a datadriven multi-task manner. [1] In this paper, we propose a Knowledge-Guided Frame Identification framework (KGFI) that integrates three types frame knowledge, including frame definitions, frame elements and frame-to-frame relations, to learn better frame representation, which guides the KGFI to jointly map target words and frames into the same embedding space and subsequently identify the best frame by calculating the dot-product similarity scores between the target word embedding and all of the frame embeddings. [2] Basically, two crucial steps are involved in CDP: The first step is to find the critical directions for each category in the embedding space, and the second step is to measure the similarity between samples and critical directions according to the projection length. [3] This paper sheds some light on the embedding spaces they create, using average cosine similarity, contextuality metrics and measures for representational similarity for comparison, revealing that BERT and NMT encoder representations look significantly different from one another. [4] First, one can hold terms constant and measure how the embedding space moves around them—much like astronomers measured the changing of celestial bodies with the seasons. [5] We generalize the color/kinematics duality of flat-space scattering amplitudes to the embedding space formulation of AdS boundary correlators. [6] Given training instances consisting of input sentences and the descriptions of their seen relations, ZS-BERT learns two functions that project sentences and relations into an embedding space by jointly minimizing the distances between them and classifying seen relations. [7] To this end, we contribute Supervised Contrastive Replay (SCR), which explicitly encourages samples from the same class to cluster tightly in embedding space while pushing those of different classes further apart during replay-based training. [8] In the second stage, we find the closest concept name representation in an embedding space to a given clinical mention. [9] Here, we present SCALEX, a deep-learning method that integrates single-cell data by projecting cells into a batch-invariant, common cell-embedding space in a truly online manner (i. [10] Deep metric learning aims at learning a mapping from the input domain to an embedding space, where semantically similar objects are located nearby and dissimilar objects far from another. [11] However, due to the various appearances of rain streaks and accumulation, it is difficult to separate rain and background information in the embedding space, which results in inaccurate deraining. [12] Additionally, we present a method for creating lexical representations that capture connotations within the embedding space and show that using the embeddings provides a statistically significant improvement on the task of stance detection when data is limited. [13] Existing approaches focus on learning an embedding space where fashion items from different categories that are visually compatible are closer to each other. [14] We train vocal and accompaniment encoders to learn a joint-embedding space of vocal and accompaniment tracks, where the embedded feature vectors of a compatible pair of vocal and accompaniment tracks lie close to each other and those of an incompatible pair lie far from each other. [15] Applying Immune2vec along with machine learning approaches to patient data exemplifies how distinct clinical conditions can be effectively stratified, indicating that the embedding space can be used for feature extraction and exploratory data analysis. [16] This paper presents an investigation of topic modeling in embedding spaces performances in the context of depression assessment. [17] The bridge nodes with small proximity and located in different communities, should be similar in embedding space since they have the same surrounding structure. [18] For each corpus, we train word embeddings, and we obtain linear transformations between the embedding spaces. [19] Given an (i) embedding space (users can choose between the predefined spaces or upload their own) and (ii) a bias specification (users can choose between existing bias specifications or create their own), DebIE can (1) compute several measures of implicit and explicit bias and modify the embedding space by executing two (mutually composable) debiasing models. [20] Loop closure candidates are found using a nearest-neighbor lookup in the embedding space. [21] We use simulated and real data to demonstrate that INSCT generates an embedding space that accurately integrates cells across experiments, platforms and species. [22] The graph entropy measurement quantifies the information of the graphs and constructs the embedding space. [23] Second, the VAE based models are not expressive enough to capture the true preference since VAE often yields an embedding space of a single modal, while in real world, user-item interactions usually exhibit multi-modality on user preference distribution. [24] However, a useful task in histopathology embedding is to train an embedding space regardless of the magnification level. [25] Our approach organizes the software requirements into clusters and sub-clusters according to requirement similarities in the embedding space. [26] Popular methods employed in network embedding either rely on implicit approximations of the principle of proximity preservation or implement it by enforcing the geometry of the embedding space, thus hindering geometric properties that networks may spontaneously exhibit. [27] Are multimodal signals useful to understand and anticipate human actions if acquired from the user viewpoint? This paper proposes to build an embedding space where inputs of different nature, but semantically correlated, are projected in a new representation space and properly exploited to anticipate the future user activity. [28] These systems consist on a feature extraction block (extracts feature maps from images) followed by a spatial dimensionality reduction block (generates compact image representations from the feature maps) and an embedding generation module (projects the image representation to the embedding space). [29] Following the mathematics allows me to define this granulated space-time as the embedding space of the skeleton of a computational spin network inside a quantum computer. [30] The key idea is to compact the bona fide speech representation and inject an angular margin to separate the spoofing attacks in the embedding space. [31] Specifically, our approach consists of three ideas: (i) extracting compression-insensitive features from both uncompressed and compressed forgeries using an adversarial learning strategy; (ii) learning a robust partition by constructing a metric loss that can reduce the distance of the paired original and compressed images in the embedding space; (iii) improving the accuracy of tampered localization with an attention-transfer module. [32] While semantic vectors were mainly used to constrain the distance between different classes in embedding space in existing ZSL methods, we first emphasize the importance of the relationship between different dimensions in semantic space. [33] This research presents a suite of data-efficient deep learning approaches to deduce the transfer function from the embedding space of English to that of Tamil, deploying three popular embedding algorithms: Word2Vec, GloVe and FastText. [34] Since the embedding distribution difference is reduced, embedding space analysis demonstrates that the model’s outputs of natural products are reliable. [35] Contrastive learning can learn better representations by keeping the instances of each class closer and pushing away instances from different classes in the embedding space concurrently. [36] Specifically, we first define the adversarial perturbations in the embedding space instead of in the discrete graph domain to circumvent the challenge of generating discrete adversarial examples. [37] For example, the overemphasis of the topological relationship by LE easily breaks the manifold structure into multiple local areas in the embedding space, which makes the spectral clustering analysis of multi-manifold data more difficult to carry out. [38] We propose an offline algorithm based on nearest-neighbor search in the embedding space to mine hard-examples from these tracks. [39] In ATCN, a novel angular distance is introduced to learn a more discriminative feature representation in the embedding space. [40] We propose a supervised local contrastive loss that leverages limited pixel-wise annotation to force pixels with the same label to gather around in the embedding space. [41] To deal with the heterogeneity of topological structure and node attributes of AHN while preserving the rich information of AHN as much as possible in embedding space, is main difficulty in AHN embedding. [42] Third, based on the improved triplet training, a deep learning model is trained to map the original samples into feature vectors in an embedding space, so that the distance between two samples in the same class is shortened whereas that of different classes is lengthened. [43] Recall that these are abstract surfaces that exist in their own right, without reference to an embedding space like \(\mathbb {R}^3\), but which nonetheless are locally homeomorphic to open sets in the plane. [44] Because the domain and global embeddings are trained with different datasets individually, their embedding spaces differ. [45] The embedding space should ideally capture the variations between all possible speakers, encoding the multiple aspects that make up speaker identity. [46] Specifically, the local relational knowledge extracted from the label-embedding space and the global relational knowledge, which is the estimated data center in the feature-embedding space, are concurrently considered to synthesize the virtual exemplars. [47] Specifically, we map the association between communities and vertices into an embedding space. [48] However, it is very hard for previous methods to discriminate the fine-grained sub-categories in the embedding space without fine-grained labels. [49] These limitations manifest as compatibility conditions that relate the quantities describing the director field to the geometry of the embedding space. [50]

convolutional neural network

The distances are computed in a feature embedding space where the scans are mapped by a convolutional neural network (CNN). [1] A multi-stream convolutional neural network (CNN) framework is proposed to effectively learn the 3D volume and 2D MIP feature vectors, respectively, and then explore their inter-dependencies in a joint volume-composition embedding space by unprojecting the 2D feature vectors into the 3D volume embedding space. [2] Second, using a convolutional neural network (CNN) and distance metric learning-based classifier, the superpixels are projected onto the embedding space and then classified by calculating the distance between superpixels’ embeddings and the centers of categories. [3] The idea of EACoupledCF is to use an outer product to explicitly model the pairwise correlations between the dimensions of the embedding space, utilize the convolutional neural network and introduce spatial attention mechanism to learn high-order features between embedded dimensions. [4] In this study, we modified the convolutional neural network to project sequences into an embedding space. [5]

cross modal retrieval

Then, armed with this dataset, we describe several approaches which leverage scene text, including a better scene-text aware cross-modal retrieval method which uses specialized representations for text from the captions and text from the visual scene, and reconcile them in a common embedding space. [1] In this paper, we propose a combination of both problems into a continual cross-modal retrieval setting, where we study how the catastrophic interference caused by new tasks impacts the embedding spaces and their cross-modal alignment required for effective retrieval. [2] For language- invariance, the framework includes cross-modal retrieval at the character level so that similar words in different languages but in the same word-embedding space are grouped together. [3]

pre trained word

Specifically, we learn an encoder to generate a debiased version of an input word embedding such that it (a) retains the semantics of the pre-trained word embedding, (b) agrees with the unbiased definition of the word according to the dictionary, and (c) remains orthogonal to the vector space spanned by any biased basis vectors in the pre-trained word embedding space. [1] The novel contributions of our solution include: (i)the introduction of a novel data representation for topic modeling based on syntactic and semantic relationships derived from distances calculated within a pre-trained word embedding space and (ii)the proposal of a new TF-IDF-based strategy, particularly developed to weight the CluWords. [2]

Joint Embedding Space

Our network consists of two components: a generator to synthesize gestures from a joint embedding space of features encoded from the input speech and the seed poses, and a discriminator to distinguish between the synthesized pose sequences and real 3D pose sequences. [1] Towards overcoming this challenge, we propose Hierarchical Moment Alignment Network (HMAN) which learns an effective joint embedding space for moments and sentences. [2] Our Siamese model comprises of a verification framework and an identification framework to learn a joint embedding space. [3] The visual semantic embedding (VSE) aims to construct a joint embedding space between visual features and semantic information, whereby classes can be well retrieved for a given image. [4] Based on this observation, we propose an Answer Category-Aware Answer Selection system (ACAAS), which jointly leverage unlabelled answer data and labelled question category data to generate answer category pseudo-labels in a joint embedding space. [5] Binaural input waveforms are mapped to a joint embedding space via a learned encoder, and separate multiplicative masking mechanisms are included for noise suppression and speaker separation. [6] Video-text retrieval is an important yet challenging task in vision-language understanding, which aims to learn a joint embedding space where related video and text instances are close to each other. [7] A common solution is to learn a joint embedding space to measure the cross-modal similarity. [8] Second, we present a Contrastive and Compatible Matching Network (C2M-Net), to directly model the relations between language queries and video moments in a joint embedding space. [9] We seamlessly combine cross-lingual pretraining objectives and cross-modal pretraining objectives in a unified framework to learn image and text in a joint embedding space from available English image-caption data, monolingual and parallel corpus. [10] To solve this problem, most approaches try to learn a joint embedding space to measure the cross-modal similarities, while paying little attention to the representation of each modality. [11] BLJAN embeds the packet’s bytes and all labels into a joint embedding space to capture their implicit correlations with a dual attention mechanism. [12] Most existing works follow a trend of learning a joint embedding space to measure the distance between global-level or local-level textual and video representation. [13] Most current methods align the proposed video moment and the text in a joint embedding space. [14] Most Zero-Shot Action Recognition (ZSAR) methods establish visual-semantic joint embedding space, which is based on commonly used visual features and semantic embeddings, to learn the correlation between actions. [15] Previous methods mitigate the modal gap by projecting sketches and images into a joint embedding space. [16] We combine two existing objective functions to make images and captions close in a joint embedding space while adapting the alignment of word embeddings between existing languages in our model. [17]

Dimensional Embedding Space

Most existing methods remove batch effects in a low-dimensional embedding space. [1] The series of adverse weather events are embedded into a high-dimensional embedding space to enable the identification of temporal event patterns similar to those observed in risky areas. [2] Most existing approaches represent graph nodes as vectors in a low-dimensional embedding space, ignoring the semantic hierarchy and the uncertainty present in the real-world graphs. [3] Then the similarity between the pair of the samples in the low-dimensional embedding space is also represented by using t-distribution to obtain the feature vectors in the embedding space from the high-dimensional data. [4] Finally, the supervised and unsupervised hypergraphs are jointed for learning optimal projection matrix and enhancing the intraclass compactness in low-dimensional embedding space. [5] Translating embedding for knowledge graphs is done with the aim of encoding structured information on entities and their rich relations in a low-dimensional embedding space. [6] This model can be also leveraged to examine the decision boundaries of the classifier in the two-dimensional embedding space. [7] The combined supervisory signals are used to learn a feature-rich and low dimensional embedding space where multiple echo cines can be temporally synchronized. [8] To realize an incremental system, we use Self-Organizing Nebulous Growths (SONG), which can incrementally construct a low-dimensional embedding space. [9] Given the fact that word embeddings can capture semantic relationship while semantically similar words tend to form semantic groups in a high-dimensional embedding space, we develop a sentence representation scheme by analyzing semantic subspaces of its constituent words. [10] The present work proposes a methodology for tackling this issue by projecting the original high-dimensional data into a low dimensional embedding space using self-organizing Kohonnen maps and later using the Weighted Multivariate FTS method (WMVFTS) for rule discovery and forecasting. [11] The trained network maps images to a high dimensional embedding space. [12] images) by calculating their distance in a high dimensional embedding space. [13] The use of a cosine distance metric and number of neighbors equal to 10 preserves both the local and global structure of the Gauguin dataset in a reduced two-dimensional embedding space thus yielding simple and clear groupings of the pigments used by the artist. [14] In this study, examining the mesoscale activity in mouse posterior cortex during a complex visual discrimination task, we found that broadly distributed choice signals defined a decision variable in a low-dimensional embedding space of multi-area activations, particularly along the ventral visual stream. [15] Today, increased attention is drawn towards network representation learning, a technique that maps nodes of a network into vectors of a low-dimensional embedding space. [16] In this study, examining the mesoscale activity in mouse posterior cortex during a complex visual discrimination task, we found that broadly distributed choice signals defined a decision variable in a low-dimensional embedding space of multi-area activations, particularly along the ventral visual stream. [17]

Word Embedding Space

Our hypothesis is that a word's representations in both word embedding spaces are more similar for non-biased words than biased words. [1] We apply an established retrofitting method to harness the verb class membership knowledge from BioVerbNet and transform a pretrained word embedding space by pulling together verbs belonging to the same semantic-syntactic class. [2] Some of these metrics are intrinsic, and are measured in word embedding spaces, and some are extrinsic, which measure the bias present downstream in the tasks that the word embeddings are plugged into. [3] This article presents a systematic literature review on quantifying the proximity between independently trained monolingual word embedding spaces. [4] In particular, VITALITY promotes serendipitous discovery of relevant literature using transformer language models, allowing users to find semantically similar papers in a word embedding space given (1) a list of input paper(s) or (2) a working abstract. [5] To study how semantic representations in cortex integrate perceptual and linguistic information, we created semantic word embedding spaces that combine models of visual and linguistic processing. [6] Bilingual lexicons map words in one language to their translations in another, and are typically induced by learning linear projections to align monolingual word embedding spaces. [7] As another kind of text data, a case study of applying it to a word embedding space is also shown. [8] Specifically, we learn an encoder to generate a debiased version of an input word embedding such that it (a) retains the semantics of the pre-trained word embedding, (b) agrees with the unbiased definition of the word according to the dictionary, and (c) remains orthogonal to the vector space spanned by any biased basis vectors in the pre-trained word embedding space. [9] The novel contributions of our solution include: (i)the introduction of a novel data representation for topic modeling based on syntactic and semantic relationships derived from distances calculated within a pre-trained word embedding space and (ii)the proposal of a new TF-IDF-based strategy, particularly developed to weight the CluWords. [10]

Latent Embedding Space

The proposed DLCL module lends on the idea of latent concepts to learn compact representations in the latent embedding space in an unsupervised way. [1] Despite successes in transferring the style based on the encoder-decoder framework, current approaches still lack the ability to preserve the content and even logic of original sentences, mainly due to the large unconstrained model space or too simplified assumptions on latent embedding space. [2] We also show that the input distribution captured in the latent embedding space can be multimodal1 for some NLP tasks, unlike those observed in computer vision tasks, and investigate if catering for the multimodal property of NLP models can improve SA analysis. [3] In this paper, we propose a novel multi-view subspace clustering method, namely Diversity and Consistency Embedding Learning (DCEL), which learns a better affinity matrix in a learned latent embedding space while simultaneously considering diversity and consistency of multi-view data. [4] Most of recent neural models have focused on how to learn sessions’ sequential representations based on the assumption that items can be projected into a single latent embedding space to describe their latent attributes. [5] To tackle this shortcoming, in this paper, we propose a new method termed Multi-view Clustering in Latent Embedding Space (MCLES), which jointly recovers a comprehensive latent embedding space, a robust global similarity matrix and an accurate cluster indicator matrix in a unified optimization framework. [6] In this paper, we propose a novel framework that models both queries and items in shared latent embedding space via a dual sequence transformer structure, which captures customer’s potential interest from the prospect of reconciling the historical queries and corresponding customers interactions. [7] We define the brain kernel in terms of a continuous nonlinear mapping from 3D brain coordinates to a latent embedding space, parametrized with a Gaussian process (GP). [8] Two learning pathways are introduced, the generation and the transformation, where the generation path focuses on learning complete representation in the latent embedding space. [9]

Common Embedding Space

Here, for this task we propose using metric learning, where a common embedding space for sessions and items is created, and distance measures dissimilarity between the provided sequence of users’ events and the next action. [1] Then, armed with this dataset, we describe several approaches which leverage scene text, including a better scene-text aware cross-modal retrieval method which uses specialized representations for text from the captions and text from the visual scene, and reconcile them in a common embedding space. [2] In the inference stage, we obtain the closest concept name representation in a common embedding space to a given mention representation. [3] Existing methods mainly focus on learning the common embedding space of images (or patches) and sentences (or words), whereby their mapping features in such embedding space can be directly measured. [4] When investigating networks over multiple time steps, a crucial question for the visualisation research community becomes apparent: How to visually trace changes of the connectivity over several transitions? Therefore, we developed an easy-to-use method that maps multiple networks to a common embedding space. [5] In this model, we construct two-branch neural networks with shared weights to map two input images into a common embedding space. [6] Most of the existing methods follow the paradigm of pair-wise learning and class-level learning to generate a common embedding space, where the similarity of heterogeneous multimodal samples can be calculated. [7] The common paradigm for the task is to first obtain fused feature of the reference image and the text, and further project them into a common embedding space with candidate images. [8]

Semantic Embedding Space

We present a method for defining warmth and competence axes in semantic embedding space, and show that the four quadrants defined by this subspace accurately represent the warmth and competence concepts, according to annotated lexicons. [1] To ensure the model's extendibility, it embeds candidate answers and recognized texts in a semantic embedding space and adopts semantic embedding based classifier to perform answer prediction. [2] We also learn an embedding of the visual channel,\ie visual features based on detected symbolism and objects, into a semantic embedding space, leveraging text semantics obtained from scene text. [3] Existing zero-shot learning (ZSL) methods usually learn a projection function between a feature space and a semantic embedding space(text or attribute space) for recognizing unseen classes. [4] The performance of classification is improved when the structure-aligned visual-semantic embedding space is transferred to the unseen classes. [5] Generally, those trackers learn a high-level semantic embedding space for feature representation but lose low-level fine-grained details. [6] Recently, many works implement zero-shot learning by transferring structural knowledge from the semantic embedding space to the image feature space. [7]

Learned Embedding Space

Then, in the domain adaptation process, beyond domain alignment, we employ Laplacian Eigenmaps to ensure the domain structure is consistently distributed in the learned embedding space. [1] It formalizes an end-to-end network architecture, referred to as b-Net, which accomplishes noise suppression through attention masking in a learned embedding space. [2] Graph nodes (superpixels) inherently belonging to the same class will be ideally clustered together in the learned embedding space. [3] The learned embedding space well captures the inherent relevance between image and text. [4] Dense retrieval (DR) has the potential to resolve the query understanding challenge in conversational search by matching in the learned embedding space. [5] The graph nodes (superpixels) inherently belonging to the same class will be ideally clustered together in the learned embedding space. [6] Architecture optimization from a learned embedding space for example through graph neural network based variational autoencoders builds a middle ground and leverages advantages from both sides. [7]

Feature Embedding Space

The distances are computed in a feature embedding space where the scans are mapped by a convolutional neural network (CNN). [1] Among different few-shot learning algorithms, metric-based methods, which focus on how to acquire a robust feature embedding space or distance metric for classification, are most studied and believed to be effective for classification tasks. [2] While such feature embedding methods have demonstrated good performance, the creation of the feature embedding space is only based on a single label, without considering label correlations in the data. [3] Final, the network was optimized by jointing Triplet Loss and Softmax Loss in the network loss layer, to obtain a feature embedding space with high intra-class compactness and inter-class separation. [4] This loss offers a more definite convergence target for each class, which not only allows the sample to be close to the homogeneous (intra-class) center but also to stay away from all heterogeneous (inter-class) centers in the feature embedding space. [5] It is designed to address a specific challenge in DA-FSL: the DA objective means that the source and target data distributions need to be aligned, typically through a shared domain-adaptive feature embedding space; but the FSL objective dictates that the target domain per class distribution must be different from that of any source domain class, meaning aligning the distributions across domains may harm the FSL performance. [6]

Shared Embedding Space

Visual-semantic embedding (VSE) networks create joint image-text representations to map images and texts in a shared embedding space to enable various information retrieval-related tasks, such as image-text retrieval, image captioning, and visual question answering. [1] More specifically, our work o↵ers (i) an approach to learn neural representations of product attribute-values within a shared embedding space as product reviews; (ii) a weighted composition strategy to develop product representations from the representation of its attributes; and, (iii) a review selection method that selects relevant reviews for the composed product representation within the neural embedding space. [2] We propose direct multimodal few-shot models that learn a shared embedding space of spoken words and images from only a few paired examples. [3] Many studies have attempted to extract representations from given entities and align them in a shared embedding space. [4] Neural models that independently project questions and answers into a shared embedding space allow for efficient continuous space retrieval from large corpora. [5] To distill the relevant knowledge and suppress the introduced noise, we design a graph alignment term in a shared embedding space between the original graph and augmented graph. [6]

Unified Embedding Space

Existing methods usually learn a similarity mapping of local parts between image and text, or embed the whole image and text into a unified embedding space. [1] The semantic-based approach maps mentions and entities in different languages to a unified embedding space, which reduces dependence on large-scale bilingual dictionaries. [2] Then, these extracted multi-typed embeddings of user and item entities are fused into the unified embedding spaces during the KG embedding process. [3] Specifically, we exploit kernel pooling to soft match topic embeddings with word and entity in a unified embedding space to generate fine-grained topical relatedness. [4]

Discriminative Embedding Space

To this end, we first learn a discriminative embedding space for vehicle-terrain interaction sounds from triplets of audio clips formed using visual features of the corresponding terrain patches and cluster the resulting embeddings. [1] In this work, we propose to seek an angularly discriminative Embedding Space where representations are continuous and have smaller intra-class angular margin and larger inter-class angular margin. [2] The latter aims at seeking appropriate distance metrics between features pairs to boost the CNNs by learning a discriminative embedding space. [3] Experimental results show that the proposed PCN can learn a more discriminative embedding space of primitives than existing self-supervised learning methods. [4]

Specific Embedding Space

Based on these modules, we propose a hierarchical attribute-aware embedding network (HAEN) which takes images and attributes as input, learns multiple attribute-specific embedding spaces, and measures fine-grained similarity in the corresponding spaces. [1] Fashion retrieval methods aim at learning a clothing-specific embedding space where images are ranked based on their global visual similarity with a given query. [2] Our idea is that we can find specific embedding space where Euclidean distance could mimic Wasserstein distance to solve the heavy computational problem. [3]

Monolingual Embedding Space

These GANs-based methods enable the alignment of two monolingual embedding spaces approximately, but the performance on the embeddings of low-frequency words (LFEs) is still unsatisfactory. [1] One mainstream method in cross-lingual word embeddings is to learn a linear mapping between two monolingual embedding spaces using a training dictionary. [2] Unsupervised cross-lingual word embedding (CLWE) methods learn a linear transformation matrix that maps two monolingual embedding spaces that are separately trained with monolingual corpora. [3]

Deep Embedding Space

We present Motion2Vec that learns a deep embedding space by minimizing a metric learning loss in a Siamese network: images from the same action segment are pulled together while being pushed away from randomly sampled images of other segments, and a time contrastive loss is used to preserve the temporal ordering of the images. [1] Metric-learning-based methods, which attempt to learn a deep embedding space on extremely large episodes, have been successfully applied to few-shot classification problems. [2] Compared with methods based on the permutation invariant training (PIT), DANs define a deep embedding space and deliver a more elaborate representation on each time-frequency (T-F) bin. [3]

Multimodal Embedding Space

, text-to-image) by jointly learning a multimodal embedding space. [1] We then integrate coarse‐grained and fine‐grained features into a multimodal embedding space, in which the similarity degrees between images and texts can be directly compared. [2] The contrastive learner interacts with the unreliable news classier to push similar credible news (or similar unreliable news) closer while moving news articles with similar content but opposite credibility labels away from each other in the multimodal embedding space. [3]

Invariant Embedding Space

Specifically, we optimize the differences between the embeddings of a support set and a query set in order to learn a channel-invariant embedding space for utterances. [1] We introduce in this letter a dual-stream mutual information distillation network (MIND-Net), whereby the non-identity specific mutual information (MI) characterized by generic face features coexistent on realistic and synthetic LR face images are distilled to render a resolution-invariant embedding space for LRFR. [2] To solve this problem, we propose the feature-level and instancelevel transfer learning in the teacher-student framework to learn a domain-invariant embedding space. [3]

Visual Embedding Space

This problem is especially pronounced in data-scarce settings where the data is relatively small (10% of the large scale MSR-VTT) to cover the rather complex audio-visual embedding space. [1] By minimizing the cross-entropy loss between a prediction and a given class label, the NN and its visual embedding space are learned to fulfill a given task. [2] In this work, we introduce Audio-Video Language Network (AVLnet), a self-supervised network that learns a shared audio-visual embedding space directly from raw video inputs. [3]

Cell Embedding Space

In this paper, we propose a deep neural network model which can embed semantic and contextual information about tabular cells in a low-dimensional cell embedding space. [1] In this paper, we propose a method to embed the semantic and contextual information about tabular cells in a low dimension cell embedding space. [2]

Contextual Embedding Space

synonyms, paraphrases), allowing translations that are different from the references, yet close in the contextual embedding space, to be treated as substantially correct. [1] We propose a novel method to estimate polysemy based on simple geometry in the contextual embedding space. [2]

Spherical Embedding Space

Besides, we use a generative directional appearance module to estimate and dynamically update the foreground/background class probabilities in a spherical embedding space. [1] Second, an effective directional appearance model-based statistics is proposed to represent the target and background on a spherical embedding space for VOS. [2]

Continuou Embedding Space

We have developed a novel encoder-decoder network termed as DVICE (Directed Variational Inference Cross Encoder), which learns a continuous embedding space to ensure better similarity learning. [1] First, we propose OptiPrompt, a novel and efficient method which directly optimizes in continuous embedding space. [2]