Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Jamie CallanSuggested from co-authorship
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works372 from public data
- Understanding bag-of-words model: a statistical framework1,392
A statistical framework which generalizes the bag-of-words representation, in which the visual words are generated by a statistical process rather than using a clustering algorithm, while the empirical performance is competitive to clustering-based method.
- 735
- 735
- SoftTriple Loss: Deep Metric Learning Without Triplet Sampling446
The SoftTriple loss is proposed to extend the SoftMax loss with multiple centers for each class, equivalent to a smoothed triplet loss where each class has a single center.
- 384
- Combining link and content for community detection380
A discriminative model for combining the link and content analysis for community detection from networked data, such as paper citation networks and Word Wide Web is proposed and introduced and hidden variables are introduced to explicitly model the popularity of nodes.
- An automatic weighting scheme for collaborative filtering371
An optimization algorithm to automatically compute the weights for different items based on their ratings from training users will create a clustered distribution for user vectors in the item space by bringing users of similar interests closer and separating users of different interests more distant.
- 362
- Extremely Low Bit Neural Network: Squeeze the Last Bit Out With ADMM307
This paper focuses on compressing and accelerating deep models with network weights represented by very small numbers of bits, referred to as extremely low bit neural network, and proposes to solve this problem using extragradient and iterative quantization algorithms that lead to considerably faster convergency compared to conventional optimization methods.
- 299
- 297
- Simple and Efficient Multiple Kernel Learning by Group Lasso295
This paper forms a closed-form solution for optimizing the kernel weights based on the equivalence between group-lasso and MKL that leads to an efficient algorithm for MKL, but also generalizes to the case for Lp-MKL.
- 282
- Online Feature Selection and Its Applications280
This article investigates the problem of online feature selection (OFS) in which an online learner is only allowed to maintain a classifier involved only a small and fixed number of features, and presents novel algorithms to solve each of the two problems.
- Large-scale text categorization by batch mode active learning277
A novel active learning algorithm that selects a batch of text documents for labeling manually in each iteration and uses the Fisher information matrix as the measurement of model uncertainty and chooses the set of documents to effectively maximize the Fisher Information matrix of a classification model.
- Multimedia Search with Pseudo-relevance Feedback263
The results are encouraging, indicating that pseudo-relevance feedback shows great promise for multimedia retrieval with very varied and errorful data.
- 249
- Online AUC Maximization241
This work develops online learning algorithm for maximizing Area Under the ROC curve (AUC), a metric that is widely used for measuring the classification performance for imbalanced data distributions, and presents two algorithms for online AUC maximization with theoretic performance guarantee.
- Learning to Rank by Optimizing NDCG Measure229
This work proposes a probabilistic framework that addresses the challenge of evaluating ranking algorithms using the expectation of NDCG over all the possible permutations of documents, and proposes an algorithm that outperforms state-of-the-art ranking algorithms on several benchmark data sets.
- 228
- Regularized Distance Metric Learning:Theory and Algorithm209
It is shown that with appropriate constraints, the generalization error of regularized distance metric learning could be independent from the dimensionality, making it suitable for handling high dimensional data.
- Approximate kernel k-means202
It is shown both analytically and empirically that the performance of approximate kernel k-means is similar to that of the kernel k -means algorithm, but with dramatically reduced run-time complexity and memory requirements.
- An Extended Level Method for Efficient Multiple Kernel Learning199
The extended level method is extended, which was originally designed for optimizing non-smooth objective functions, to convex-concave optimization, and applies it to multiple kernel learning, and overcomes the drawbacks of SILP and SD.
- One-pass AUC optimization198
A regression-based algorithm is developed that requires going through the training data only once without storing the entire training dataset, where conventional online learning algorithms cannot be applied directly because AUC is measured by a sum of losses defined over pairs of instances from different classes.
- Beyond Appearance: A Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks192
This paper aims to learn a general human representation from massive unlabeled human images which can benefit downstream human-centric tasks to the maximum extent using a Semantic cOntrollable seLf-supervIseD lEaRning framework.
- 190
- Zen-NAS: A Zero-Shot NAS for High-Performance Image Recognition186
A novel zero-shot index dubbed Zen- score is proposed to rank the architectures in Neural Architecture Search by maximizing the Zen-Score of the target network under given inference budgets, which is magnitude times faster on multiple server-side and mobile-side GPU platforms with state-of-the-art accuracy on ImageNet.
- A Boosting Framework for Visuality-Preserving Distance Metric Learning and Its Application to Medical Image Retrieval184
This work presents a boosting framework for distance metric learning that aims to preserve both visual and semantic similarities and shows that the boosting framework compares favorably to state-of-the-art approaches fordistance metric learning in retrieval accuracy, with much lower computational cost.
- Semisupervised SVM batch mode active learning with applications to image retrieval182
This work proposes a novel scheme for explicitly addressing the drawbacks of conventional SVM active learning, which first learns a kernel function from a mixture of labeled and unlabeled data, and therefore alleviates the problem of small-sized training data.
- On the Equivalence of Nonnegative Matrix Factorization and K-means - Spectral Clustering180
A systematic analysis of nonnegative matrix factorization (NMF) relating to data cluster- ing and the importance of orthogonality in NMF and soft clustering nature of NMF is emphasized.
- A unified log-based relevance feedback scheme for image retrieval173
A unified framework for log-based relevance feedback that integrates the log of feedback data into the traditional relevance feedback schemes to learn effectively the correlation between low-level image features and high-level concepts is proposed.
- A language modeling framework for resource selection and results merging165
This paper extends the language modeling approach to integrate resource selection, ad-hoc searching, and merging of results from different text databases into a single probabilistic retrieval model, designed primarily for Intranet environments.
- KVT: k-NN Attention for Boosting Vision Transformers159
A sparse attention scheme, dubbed k-NN attention, which naturally inherits the local bias of CNNs without introducing convolutional operations, and allows for the exploration of long range correlation and filter out irrelevant tokens by choosing the most similar tokens from the entire image.
- Linear Convergence with Condition Number Independent Access of Full Gradients158
This paper proposes to remove the dependence on the condition number by allowing the algorithm to access stochastic gradients of the objective function, and presents a novel algorithm named Epoch Mixed Gradient Descent (EMGD) that is able to utilize two kinds of gradients.
- 158
- 157
- 157
- Effective automatic image annotation via a coherent language model and active learning153
A coherent language model for automatic image annotation is proposed that takes into account the word-to-word correlation by estimating a coherent language models for an image to significantly reduce the required number of annotated image examples.
- Negative pseudo-relevance feedback in content-based video retrieval148
This work presents a novel approach that uses pseudo-relevance feedback from retrieved items that are NOT similar to the query items without further inquiring user feedback and suggests a score combination scheme via posterior probability estimation.
- Learning nonparametric kernel matrices from pairwise constraints142
This paper forms the problem into Semi-Definite Programs (SDP), and proposes an efficient algorithm to solve the SDP problem by introducing the graph Laplacian of the observed data as a regularizer when optimizing the kernel matrix with respect to the pairwise constraints.
- Hybrid Relation Guided Set Matching for Few-shot Action Recognition138
A novel Hybrid Relation guided Set Matching (HyRSM) approach that incorporates two key components: hybrid relation module and set matching metric that can be highly informative and flexible to predict query categories under the few-shot settings is proposed.
- 136
- Learning Position and Target Consistency for Memory-based Video Object Segmentation134
LCM introduces an object-level relationship from the target to maintain target consistency, making LCM more robust to error drifting, and achieves state-of-the-art performance on both DAVIS and Youtube-VOS benchmark.
- 131
- 131
- 130
- 123
- Collaborative filtering with decoupled models for preferences and ratings120
This work builds two separate models for collaborative filtering: a preference model capturing which items are favored by the user and a rating model capturing how the user would rate an item given the preference information.
- Exploitation and exploration in a performance based contextual advertising system112
This paper develops two novel EE strategies for online advertising that can adaptively balance the two aspects of EE by automatically learning the optimal tradeoff and incorporating confidence metrics of historical performance.
- 111
- 110
- Efficient distance metric learning by adaptive sampling and mini-batch stochastic gradient descent (SGD)107
This work develops hybrid approaches that combine the strength of adaptive sampling with that of mini-batch online learning techniques to further improve the computational efficiency of SGD for DML and proves the theoretical guarantees for both adaptive sampling and mini- batch based approaches for D ML.
- Online feature selection for mining big data107
This work investigates the problem of Online Feature Selection (OFS) in which the online learner is only allowed to maintain a classifier involved a small and fixed number of features, and presents an effective algorithm to solve the problem, and gives the theoretical analysis and empirical performance of the proposed algorithms.
- 107
- 105
- CHEX: CHannel EXploration for CNN Model Compression104
This paper proposes to repeatedly prune and regrow the channels throughout the training process, which reduces the risk of pruning important channels prematurely, and can effectively reduce the FLOPs of diverse CNN architectures on a variety of computer vision tasks.
- 99
- Double Updating Online Learning98
An extensive set of empirical evaluations for both binary and multi-class online learning tasks show that the proposed DUOL technique is considerably more effective than the state-of-the-art online learning algorithms.
- Online Multiple Kernel Classification97
This work addresses a new research problem, termed Online Multiple Kernel Classification (OMKC), which learns a kernel-based prediction function by selecting a subset of predefined kernel functions in an online learning fashion, and develops stochastic selection strategies that randomly select a portion of kernels for combination and model updating, thus improving the learning efficiency.
- Distance metric learning from uncertain side information with application to automated photo tagging97
A novel probabilistic distance metric learning scheme is proposed that automatically derives constraints from the uncertain side information, and efficiently learns a distance metric from the derived constraints.
- Title language model for information retrieval96
In the experiments with four different TREC document collections, the title language model for information retrieval with the new smoothing method outperforms both the traditional language model and the vector space model for IR significantly.
- Online Multiple Kernel Similarity Learning for Visual Search95
This paper investigates an online kernel similarity learning framework for learning kernel-based proximity functions which goes beyond the conventional linear distance metric learning approaches and proposes a novel online multiple kernel similarity (OMKS) learning method which learns a flexible nonlinear proximity function with multiple kernels to improve visual similarity search in CBIR.
- 91
- Batch Mode Active Learning with Applications to Text Categorization and Image Retrieval90
A framework for batch mode active learning, which selects a number of informative examples for manual labeling in each iteration to reduce the redundancy among the selected examples such that each example provides unique information for model updating.
- Semi-Crowdsourced Clustering: Generalizing Crowd Labeling by Robust Distance Metric Learning88
A new approach for clustering is proposed, called semi-crowdsourced clustering that effectively combines the low-level features of objects with the manual annotations of a subset of the objects obtained via crowdsourcing and outperforms state-of-the-art distance metric learning algorithms in both clustering accuracy and computational efficiency.
- 84
- Web image retrieval re-ranking with relevance model84
A re-ranking method to improve Web image retrieval by reordering the images retrieved from an image search engine based on a relevance model, which is a probabilistic model that evaluates the relevance of the HTML document linking to the image, and assigns a probability of relevance.
- 83
- Online Multiple Kernel Learning: Algorithms and Mistake Bounds83
This paper introduces a new research problem, termed Online Multiple Kernel Learning (OMKL), that aims to learn a kernel based prediction function from a pool of predefined kernels in an online learning fashion, and proposes both deterministic and stochastic approaches in the two setups for OMKL.
- DR Loss: Improving Object Detection by Distributional Ranking82
This work proposes a novel distributional ranking (DR) loss, which mitigates the intra-class imbalance issue in background candidates but also improves the efficiency for the ranking algorithm.
- Robust Optimization over Multiple Domains80
This work proposes a framework of robust optimization over multiple domains and proposes to learn the model and the adversarial distribution simultaneously with the stochastic algorithm for efficiency, and analyzes the convergence rate for convex and non-convex models.
- Non-monotonic feature selection79
This work develops an algorithm for non-monotonic feature selection that approximates the related combinatorial optimization problem by a Multiple Kernel Learning (MKL) problem, and presents a strategy that derives a discrete solution from the approximate solution of MKL, and shows the performance guarantee.
- On the Use of Dynamic Bayesian Networks in Reconstructing Functional Neuronal Networks from Spike Train Ensembles78
It is demonstrated that DBNs can infer the underlying nonlinear and time-varying causal interactions between these neurons and can discriminate between mono- and polysynaptic links between them under certain constraints governing their putative connectivity.
- 76
- BoostCluster72
The proposed boosting framework, branded as BoostCluster, is able to iteratively improve the accuracy of any given clustering algorithm by exploiting the pairwise constraints and is comparable to the state-of-the-art algorithms for data clustering with side information.
- 71
- Efficient Convex Relaxation for Transductive Support Vector Machine70
This work proposes solving Transductive SVM via a convex relaxation, which converts the NP-hard problem to a semi-definite programming, and shows the promising performance of the proposed algorithm in comparison with other state-of-the-art implementations of Transductives SVM.
- 68
- 67
- Discourse Structure for Context Question Answering67
The role of discourse modeling in context question answering is discussed, a semantic-rich discourse representation is motivated and the impact of refined discourse structure on question answering are discussed.
- Semi-Supervised Boosting for Multi-Class Classification66
A semi-supervised boosting framework, termed Multi-Class Semi-Supervised Boosting (MCSSB) is proposed, that directly solves the semi- supervised multi-class learning problem and exploits both classification confidence and similarities among examples when deciding the pseudo-labels for unlabeled examples.
- 64
- 64
- Stochastic Gradient Descent with Only One Projection64
This work develops novel stochastic optimization algorithms that do not need intermediate projections to obtain a feasible solution in the given domain and achieves an O(1/√T) convergence rate for general convex optimization and a O(ln T/T) rate for strongly conveX optimization under mild conditions about the domain and the objective function.
- Multi-label Multiple Kernel Learning64
The proposed learning formulation leads to a non-smooth min-max problem, which can be cast into a semi-infinite linear program (SILP) and an approximate formulation with a guaranteed error bound which involves an unconstrained convex optimization problem.
- Self-supervised Video Representation Learning by Context and Motion Decoupling62
This work develops a method that explicitly decouples motion supervision from context bias through a carefully designed pretext task that improves the quality of the learned video representation and finds the motion prediction to be a strong regularization for video networks.
- 61
- Tattoo-ID: Automatic Tattoo Image Retrieval for Suspect and Victim Identification61
This work examines several key design issues related to building a prototype CBIR system for tattoo image database and computes the similarity between the query and stored tattoos based on image content to retrieve the most similar tattoos.
- Collaborative image retrieval via regularized metric learning60
This paper presents a novel metric learning approach, named “regularized metric learning,” for collaborative image retrieval, which learns a distance metric by exploring the correlation between low-level image features and the log data of users' relevance judgments.
- Scaled ReLU Matters for Training Vision Transformers59
It is verified, both theoretically and empirically, that scaled ReLU in conv-stem not only improves training stabilization, but also increases the diversity of patch tokens, thus boosting peak performance with a large margin via adding few parameters and flops.
- 59
- A learning framework using Green's function and kernel regularization with application to recommender system58
Green's function provides a well-defined distance metric on a generic weighted graph, either as the effective distance on the network of electric resistors, or the average commute time in random walks, and it is shown that for unsupervised learning this approach is identical to Ratio Cut and Normalized Cut spectral clustering algorithms.
- User term feedback in interactive text-based image retrieval58
The results indicate that, although term feedback has a positive effect by allowing users to identify more relevant terms, it also has a strong negative effect by providing more opportunities for users to specify irrelevant terms.
- Video retrieval using speech and image information58
An evaluation on the effects of different types of information used for video retrieval from a video collection found image matching and video OCR proved to be the deciding aspects of video information retrieval.
- Making Vision Transformers Efficient from A Token Sparsification View56
A novel Semantic Token ViT (STViT) is proposed, for efficient global and local vision transformers, which can also be revised to serve as backbone for downstream tasks, and a STViT-R(ecovery) network is designed to restore the detailed spatial information based on the ST ViT, making it work for upstream tasks, which is powerless for previous token sparsification methods.
- 56
- 55
- A Bayesian Approach Toward Finding Communities and Their Evolutions in Dynamic Social Networks55
This paper proposes a dynamic stochastic block model for finding communities and their evolutions in a dynamic social network that captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network.
- Localized Support Vector Machine and Its Efficient Algorithm54
This paper presents a framework called Localized Support Vector Machine (LSVM), which builds multiple linear SVM models from the training data, which achieves comparable accuracy as LSVM but with significant computational savings.
- 53
- Learning Bregman Distance Functions and Its Application for Semi-Supervised Clustering53
This paper proposes a novel scheme that learns nonlinear Bregman distance functions from side information using a non-parametric approach that is similar to support vector machines and presents an efficient learning algorithm for the proposed scheme.
- 52
- TransFGU: A Top-Down Approach to Fine-Grained Unsupervised Semantic Segmentation51
This work proposes the first top-down unsupervised semantic segmentation framework for fine-grained segmentation in extremely complicated scenarios and shows that it is robust to both object-centric and scene-centric datasets under different semantic granularity levels, and outperforms all the current state-of-the-art bottom-up methods.
- Decoupling and Recoupling Spatiotemporal Representation for RGB-D-based Motion Recognition51
Seamless combination of these novel designs forms a robust spatiotemporal representation and achieves better performance than state-of-the-art methods on four public motion datasets.
- 51
- Semi-supervised learning by mixed label propagation51
This paper proposes a new graph-based approach, termed as "mixed label propagation" which is able to effectively explore both similarity and dissimilarity simultaneously, and determines the assignment of class labels by minimizing the energy function associated with positive similarity, and maximizing the energyfunction associated with negative similarity.
- Image Tag Completion by Noisy Matrix Recovery49
This study focuses on the problem of tag completion that aims to simultaneously enrich the missing tags and remove noisy tags and proposes a noisy matrix recovery algorithm that outperforms state-of-the-art approaches in terms of both effectiveness and efficiency.
- Directed Network Community Detection: A Popularity and Productivity Link Model49
A probabilistic model for directed network community detection that aims to model both incoming links and outgoing links simultaneously and differentially, and introduces latent variables node productivity and node popularity to explicitly capture outgoing links and incoming links, respectively.
- Ranking refinement and its application to information retrieval49
This work presents a novel boosting framework for ranking refinement that can effectively leverage the uses of the two sources of information and significantly outperforms the baseline algorithms that incorporate the outputs from the base ranker as an additional feature.
- 48
- An Information Geometry Approach for Distance Metric Learning46
This paper presents two metric learning algorithms, one for linear distance metric and one for nonlinear distance using a kernel function, which have a closed-form solution and are computationally more efficient.
- A maximum coherence model for dictionary-based cross-language information retrieval45
A novel statistical model is proposed, named ``maximum coherence model'', which estimates the translation probabilities of query words that are consistent with the word co-occurrence statistics, and is able to estimate translations of multiple query words simultaneously.
- 44
- Semi-supervised ensemble ranking42
The proposed SSER algorithm is formulated as an SVM-like quadratic program (QP), and therefore can be solved efficiently by taking advantage of optimization techniques that were widely used in existing SVM solvers.
- 41
- 41
- Boosting multi-kernel locality-sensitive hashing for scalable image retrieval41
A novel Boosting Multi-Kernel Locality-Sensitive Hashing (BMKLSH) scheme is proposed that significantly boosts the retrieval performance of KLSH by making use of multiple kernels and outperforms the state-of-the-art techniques.
- Semi-supervised Clustering by Input Pattern Assisted Pairwise Similarity Matrix Completion40
This work develops a framework for semisupervised clustering based on input pattern assisted matrix completion, and shows that under appropriate conditions, only O(log n) pairwise constraints are needed to accurately recover the true cluster partition.
- Automated annotation of Drosophila gene expression patterns using a controlled vocabulary40
A novel computational framework for annotating gene expression patterns using a controlled vocabulary is presented, and efficient convex formulations to integrate the kernels derived from various features are proposed to exploit the complementary information conveyed by different features and incorporate the correlation among patterns sharing common structures.
- Meta-scoring40
A meta-scoring function, which takes as input the document-term matrix generated by some term weighting scheme and computes a goodness score from the documents generated, can be a substitute for the precision-recall measurement that needs relevance judgments of human subject.
- Automatic title generation for spoken broadcast news40
The results show that title generation for speech recognized news documents is possible at a level approaching the accuracy of titles generated for perfect text transcriptions.
- Multi-label Multiple Kernel Learning by Stochastic Approximation: Application to Visual Object Recognition38
This work develops an efficient algorithm for multi-label multiple kernel learning (ML-MKL) that combines the worst-case analysis with stochastic approximation and shows that the complexity of the algorithm is O(m1/3√lnm), where m is the number of classes.
- 36
- 35
- 34
- Self-supervised Motion Learning from Static Images33
The model learns to encode motion information by classifying pseudo motions generated by MoSI, and introduces a static mask in pseudo motions to create local motion patterns, which forces the model to additionally locate notable motion areas for the correct classification.
- 31
- 30
- Smooth Optimization for Effective Multiple Kernel Learning30
This work transforms the optimization of multiple Kernel Learning into a single smooth convex optimization problem and employs the Nesterov’s method to efficiently solve the optimization problem.
- 29
- Unified filtering by combining collaborative filtering and content-based filtering via mixture model and exponential model29
A unified probabilistic framework that allows the mutual interaction between content information and rating information in an integrated way and outperforms a pure collaborative filtering approach, a pure content-based filtering approach and another unified filtering algorithm is presented.
- 28
- 28
- Achieving Human Parity on Visual Question Answering27
A novel hierarchical integration of vision and language AliceMind-MMU (ALIbaba’s Collection of Encoder-decoders from Machine IntelligeNce lab of Damo academy - MultiMedia Understanding) is introduced, which leads to similar or even slightly better results than a human being does on VQA.
- 27
- 27
- 27
- 26
- 26
- 26
- 25
- 25
- 25
- 25
- Learning from Untrimmed Videos: Self-Supervised Video Representation Learning with Hierarchical Consistency24
A hierarchical consistency learning framework HiCo is presented, and it is shown that not only HiCo can generate stronger representations on untrimmed videos, it also improves the representation quality when applied to trimmed videos, in contrast to standard contrastive learning that fails to learn appropriate representations from untrimed videos.
- 24
- 24
- An Empirical Study on Distribution Shift Robustness From the Perspective of Pre-Training and Data Augmentation23
This paper studies the distribution shift problem from the perspective of pre-training and data augmentation, two important factors in the practice of deep learning that have not been systematically investigated by existing work.
- 23
- 23
- 22
- 22
- 22
- 22
- 22
- DUOL: A Double Updating Approach for Online Learning22
It is shown that the mistake bound can be significantly improved by the proposed online learning method, termed Double Updating Online Learning, or DUOL for short, which is in general considerably more effective than the state-of-the-art online learning algorithms.
- 22
- 21
- 21
- 21
- 20
- 19
- 18
- 18
- 18
- 18
- 18
- 18
- 18
- 17
- 17
- 17
- 17
- 17
- 17
- 17
- 16
- 16
- 16
- 15
- 15
- 15
- 15
- 15
- 15
- 15
- 14
- 14
- 14
- 14
- Machine learning for information retrieval14
This tutorial will benefit a large body of audience in the information retrieval community, ranging from students who are new to machine learning to the seasoned researchers who would like to understand the recent advance in machine learning for information retrieval research.
- 13
- 13
- 13
- 13
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- Discriminative Cluster Refinement: Improving Object Category Recognition Given Limited Training Data12
- 11
- 11
- 11
- 11
- 11
- 10
- 10
- 10
- 10
- 10
- 10
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 7
- Proceedings of the first SIGMM workshop on Social media7
The program includes a keynote speech by Ramesh Jain on "Influence of Multimedia in Emerging Social Web Systems", and a panel discussion on "Emerging Challenges in Social Media Research".
- Language model for IR using collection information7
A new language model is presented that is able to take advantage of the category information for documents to improve the retrieval accuracy and outperforms the traditional language model over the TREC4 dataset.
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 3
- ParamCrop: Parametric Cubic Cropping for Video Contrastive Learning3
It is shown that adaptively controlling the disparity between two augmented views along the training process enhances the quality of the learned representations, and this adaptive and gradual increase in the disparity yielded by ParamCrop is beneficial to learning a strong and generalized representation for downstream tasks.
- 3
- 3
- Train a One-Million-Way Instance Classifier for Unsupervised Visual Representation Learning3
This paper presents a simple unsupervised visual representation learning method with a pretext task of discriminating all images in a dataset using a parametric, instance-level classifier, and introduces a hybrid parallel training framework to make large-scale training feasible.
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Proceedings of second ACM SIGMM workshop on Social media2
The program includes two keynote speeches, one by Ramesh Sarukkai on "Creating the Perfect Social Video Stream", and the other by Daniel Gatica- Perez on "The You in YouTube".
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- Enhancing semi‐supervised contrastive learning through saliency map for diabetic retinopathy grading1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.