PyTorchKR Papers

최신 논문

총 11,878편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2306.16122

Semantic Positive Pairs for Enhancing Visual Representation Learning of Instance Discrimination methods

Self-supervised learning algorithms (SSL) based on instance discrimination have shown promising results, performing competitively or even outperforming supervised learning counterparts in some downstream tasks. Such approaches employ data augmentation to create two views of the same instance (i.e., positive pairs) and encourage the model to learn good representations by attracting these views closer in the embedding space without collapsing to the trivial solution. However, data augmentation is limited in representing positive pairs, and the repulsion process between the instances during contrastive learning may discard important features for instances that have similar categories. To address this issue, we propose an approach to identify those images with similar semantic content and treat them as positive instances, thereby reducing the chance of discarding important features during representation learning and increasing the richness of the latent representation. Our approach is generic and could work with any self-supervised instance discrimination frameworks such as MoCo and SimSiam. To evaluate our method, we run experiments on three benchmark datasets: ImageNet, STL-10 and CIFAR-10 with different instance discrimination SSL approaches. The experimental results show that our approach consistently outperforms the baseline methods across all three datasets; for instance, we improve upon the vanilla MoCo-v2 by 4.1% on ImageNet under a linear evaluation protocol over 800 epochs. We also report results on semi-supervised learning, transfer learning on downstream tasks, and object detection.

arXiv:2306.15925공식 코드 · Papers with Code

Subclass-balancing Contrastive Learning for Long-tailed Recognition

Long-tailed recognition with imbalanced class distribution naturally emerges in practical machine learning applications. Existing methods such as data reweighing, resampling, and supervised contrastive learning enforce the class balance with a price of introducing imbalance between instances of head class and tail class, which may ignore the underlying rich semantic substructures of the former and exaggerate the biases in the latter. We overcome these drawbacks by a novel ``subclass-balancing contrastive learning (SBCL)'' approach that clusters each head class into multiple subclasses of similar sizes as the tail classes and enforce representations to capture the two-layer class hierarchy between the original classes and their subclasses. Since the clustering is conducted in the representation space and updated during the course of training, the subclass labels preserve the semantic substructures of head classes. Meanwhile, it does not overemphasize tail class samples, so each individual instance contribute to the representation learning equally. Hence, our method achieves both the instance- and subclass-balance, while the original class labels are also learned through contrastive learning among subclasses from different classes. We evaluate SBCL over a list of long-tailed benchmark datasets and it achieves the state-of-the-art performance. In addition, we present extensive analyses and ablation studies of SBCL to verify its advantages.

arXiv:2306.15876

Hybrid Distillation: Connecting Masked Autoencoders with Contrastive Learners

Representation learning has been evolving from traditional supervised training to Contrastive Learning (CL) and Masked Image Modeling (MIM). Previous works have demonstrated their pros and cons in specific scenarios, i.e., CL and supervised pre-training excel at capturing longer-range global patterns and enabling better feature discrimination, while MIM can introduce more local and diverse attention across all transformer layers. In this paper, we explore how to obtain a model that combines their strengths. We start by examining previous feature distillation and mask feature reconstruction methods and identify their limitations. We find that their increasing diversity mainly derives from the asymmetric designs, but these designs may in turn compromise the discrimination ability. In order to better obtain both discrimination and diversity, we propose a simple but effective Hybrid Distillation strategy, which utilizes both the supervised/CL teacher and the MIM teacher to jointly guide the student model. Hybrid Distill imitates the token relations of the MIM teacher to alleviate attention collapse, as well as distills the feature maps of the supervised/CL teacher to enable discrimination. Furthermore, a progressive redundant token masking strategy is also utilized to reduce the distilling costs and avoid falling into local optima. Experiment results prove that Hybrid Distill can achieve superior performance on different benchmarks.

arXiv:2306.15836

A generic self-supervised learning (SSL) framework for representation learning from spectra-spatial feature of unlabeled remote sensing imagery

Remote sensing data has been widely used for various Earth Observation (EO) missions such as land use and cover classification, weather forecasting, agricultural management, and environmental monitoring. Most existing remote sensing data-based models are based on supervised learning that requires large and representative human-labelled data for model training, which is costly and time-consuming. Recently, self-supervised learning (SSL) enables the models to learn a representation from orders of magnitude more unlabelled data. This representation has been proven to boost the performance of downstream tasks and has potential for remote sensing applications. The success of SSL is heavily dependent on a pre-designed pretext task, which introduces an inductive bias into the model from a large amount of unlabelled data. Since remote sensing imagery has rich spectral information beyond the standard RGB colour space, the pretext tasks established in computer vision based on RGB images may not be straightforward to be extended to the multi/hyperspectral domain. To address this challenge, this work has designed a novel SSL framework that is capable of learning representation from both spectra-spatial information of unlabelled data. The framework contains two novel pretext tasks for object-based and pixel-based remote sensing data analysis methods, respectively. Through two typical downstream tasks evaluation (a multi-label land cover classification task on Sentienl-2 multispectral datasets and a ground soil parameter retrieval task on hyperspectral datasets), the results demonstrate that the representation obtained through the proposed SSL achieved a significant improvement in model performance.

arXiv:2306.15796

ConKI: Contrastive Knowledge Injection for Multimodal Sentiment Analysis

Multimodal Sentiment Analysis leverages multimodal signals to detect the sentiment of a speaker. Previous approaches concentrate on performing multimodal fusion and representation learning based on general knowledge obtained from pretrained models, which neglects the effect of domain-specific knowledge. In this paper, we propose Contrastive Knowledge Injection (ConKI) for multimodal sentiment analysis, where specific-knowledge representations for each modality can be learned together with general knowledge representations via knowledge injection based on an adapter architecture. In addition, ConKI uses a hierarchical contrastive learning procedure performed between knowledge types within every single modality, across modalities within each sample, and across samples to facilitate the effective learning of the proposed representations, hence improving multimodal sentiment predictions. The experiments on three popular multimodal sentiment analysis benchmarks show that ConKI outperforms all prior methods on a variety of performance metrics.

arXiv:2306.15711공식 코드 · Papers with Code

Semi-supervised Multimodal Representation Learning through a Global Workspace

Recent deep learning models can efficiently combine inputs from different modalities (e.g., images and text) and learn to align their latent representations, or to translate signals from one domain to another (as in image captioning, or text-to-image generation). However, current approaches mainly rely on brute-force supervised training over large multimodal datasets. In contrast, humans (and other animals) can learn useful multimodal representations from only sparse experience with matched cross-modal data. Here we evaluate the capabilities of a neural network architecture inspired by the cognitive notion of a "Global Workspace": a shared representation for two (or more) input modalities. Each modality is processed by a specialized system (pretrained on unimodal data, and subsequently frozen). The corresponding latent representations are then encoded to and decoded from a single shared workspace. Importantly, this architecture is amenable to self-supervised training via cycle-consistency: encoding-decoding sequences should approximate the identity function. For various pairings of vision-language modalities and across two datasets of varying complexity, we show that such an architecture can be trained to align and translate between two modalities with very little need for matched data (from 4 to 7 times less than a fully supervised approach). The global workspace representation can be used advantageously for downstream classification tasks and for robust transfer learning. Ablation studies reveal that both the shared workspace and the self-supervised cycle-consistency training are critical to the system's performance.

arXiv:1808.35573

MDGCF: Multi-Dependency Graph Collaborative Filtering with Neighborhood- and Homogeneous-level Dependencies

Due to the success of graph convolutional networks (GCNs) in effectively extracting features in non-Euclidean spaces, GCNs has become the rising star in implicit collaborative filtering. Existing works, while encouraging, typically adopt simple aggregation operation on the user-item bipartite graph to model user and item representations, but neglect to mine the sufficient dependencies between nodes, e.g., the relationships between users/items and their neighbors (or congeners), resulting in inadequate graph representation learning. To address these problems, we propose a novel Multi-Dependency Graph Collaborative Filtering (MDGCF) model, which mines the neighborhood- and homogeneous-level dependencies to enhance the representation power of graph-based CF models. Specifically, for neighborhood-level dependencies, we explicitly consider both popularity score and preference correlation by designing a joint neighborhood-level dependency weight, based on which we construct a neighborhood-level dependencies graph to capture higher-order interaction features. Besides, by adaptively mining the homogeneous-level dependencies among users and items, we construct two homogeneous graphs, based on which we further aggregate features from homogeneous users and items to supplement their representations, respectively. Extensive experiments on three real-world benchmark datasets demonstrate the effectiveness of the proposed MDGCF. Further experiments reveal that our model can capture rich dependencies between nodes for explaining user behaviors.

arXiv:2306.16326

Representation Learning via Variational Bayesian Networks

We present Variational Bayesian Network (VBN) - a novel Bayesian entity representation learning model that utilizes hierarchical and relational side information and is particularly useful for modeling entities in the ``long-tail'', where the data is scarce. VBN provides better modeling for long-tail entities via two complementary mechanisms: First, VBN employs informative hierarchical priors that enable information propagation between entities sharing common ancestors. Additionally, VBN models explicit relations between entities that enforce complementary structure and consistency, guiding the learned representations towards a more meaningful arrangement in space. Second, VBN represents entities by densities (rather than vectors), hence modeling uncertainty that plays a complementary role in coping with data scarcity. Finally, we propose a scalable Variational Bayes optimization algorithm that enables fast approximate Bayesian inference. We evaluate the effectiveness of VBN on linguistic, recommendations, and medical inference tasks. Our findings show that VBN outperforms other existing methods across multiple datasets, and especially in the long-tail.

arXiv:2306.16195공식 코드 · Papers with Code

Enhancing Dialogue Generation via Dynamic Graph Knowledge Aggregation

Incorporating external graph knowledge into neural chatbot models has been proven effective for enhancing dialogue generation. However, in conventional graph neural networks (GNNs), message passing on a graph is independent from text, resulting in the graph representation hidden space differing from that of the text. This training regime of existing models therefore leads to a semantic gap between graph knowledge and text. In this study, we propose a novel framework for knowledge graph enhanced dialogue generation. We dynamically construct a multi-hop knowledge graph with pseudo nodes to involve the language model in feature aggregation within the graph at all steps. To avoid the semantic biases caused by learning on vanilla subgraphs, the proposed framework applies hierarchical graph attention to aggregate graph features on pseudo nodes and then attains a global feature. Therefore, the framework can better utilise the heterogeneous features from both the post and external graph knowledge. Extensive experiments demonstrate that our framework outperforms state-of-the-art (SOTA) baselines on dialogue generation. Further analysis also shows that our representation learning framework can fill the semantic gap by coagulating representations of both text and graph knowledge. Moreover, the language model also learns how to better select knowledge triples for a more informative response via exploiting subgraph patterns within our feature aggregation process. Our code and resources are available at https://github.com/tangg555/SaBART.

arXiv:2306.15938

Interpretable Anomaly Detection in Cellular Networks by Learning Concepts in Variational Autoencoders

This paper addresses the challenges of detecting anomalies in cellular networks in an interpretable way and proposes a new approach using variational autoencoders (VAEs) that learn interpretable representations of the latent space for each Key Performance Indicator (KPI) in the dataset. This enables the detection of anomalies based on reconstruction loss and Z-scores. We ensure the interpretability of the anomalies via additional information centroids (c) using the K-means algorithm to enhance representation learning. We evaluate the performance of the model by analyzing patterns in the latent dimension for specific KPIs and thereby demonstrate the interpretability and anomalies. The proposed framework offers a faster and autonomous solution for detecting anomalies in cellular networks and showcases the potential of deep learning-based algorithms in handling big data.

arXiv:2306.15651

Dental CLAIRES: Contrastive LAnguage Image REtrieval Search for Dental Research

Learning about diagnostic features and related clinical information from dental radiographs is important for dental research. However, the lack of expert-annotated data and convenient search tools poses challenges. Our primary objective is to design a search tool that uses a user's query for oral-related research. The proposed framework, Contrastive LAnguage Image REtrieval Search for dental research, Dental CLAIRES, utilizes periapical radiographs and associated clinical details such as periodontal diagnosis, demographic information to retrieve the best-matched images based on the text query. We applied a contrastive representation learning method to find images described by the user's text by maximizing the similarity score of positive pairs (true pairs) and minimizing the score of negative pairs (random pairs). Our model achieved a hit@3 ratio of 96% and a Mean Reciprocal Rank (MRR) of 0.82. We also designed a graphical user interface that allows researchers to verify the model's performance with interactions.

arXiv:2306.15661

Enhancing Representation Learning on High-Dimensional, Small-Size Tabular Data: A Divide and Conquer Method with Ensembled VAEs

Variational Autoencoders and their many variants have displayed impressive ability to perform dimensionality reduction, often achieving state-of-the-art performance. Many current methods however, struggle to learn good representations in High Dimensional, Low Sample Size (HDLSS) tasks, which is an inherently challenging setting. We address this challenge by using an ensemble of lightweight VAEs to learn posteriors over subsets of the feature-space, which get aggregated into a joint posterior in a novel divide-and-conquer approach. Specifically, we present an alternative factorisation of the joint posterior that induces a form of implicit data augmentation that yields greater sample efficiency. Through a series of experiments on eight real-world datasets, we show that our method learns better latent representations in HDLSS settings, which leads to higher accuracy in a downstream classification task. Furthermore, we verify that our approach has a positive effect on disentanglement and achieves a lower estimated Total Correlation on learnt representations. Finally, we show that our approach is robust to partial features at inference, exhibiting little performance degradation even with most features missing.

arXiv:2306.15128공식 코드 · Papers with Code

MIMIC: Masked Image Modeling with Image Correspondences

Dense pixel-specific representation learning at scale has been bottlenecked due to the unavailability of large-scale multi-view datasets. Current methods for building effective pretraining datasets heavily rely on annotated 3D meshes, point clouds, and camera parameters from simulated environments, preventing them from building datasets from real-world data sources where such metadata is lacking. We propose a pretraining dataset-curation approach that does not require any additional annotations. Our method allows us to generate multi-view datasets from both real-world videos and simulated environments at scale. Specifically, we experiment with two scales: MIMIC-1M with 1.3M and MIMIC-3M with 3.1M multi-view image pairs. We train multiple models with different masked image modeling objectives to showcase the following findings: Representations trained on our automatically generated MIMIC-3M outperform those learned from expensive crowdsourced datasets (ImageNet-1K) and those learned from synthetic environments (MULTIVIEW-HABITAT) on two dense geometric tasks: depth estimation on NYUv2 (1.7%), and surface normals estimation on Taskonomy (2.05%). For dense tasks which also require object understanding, we outperform MULTIVIEW-HABITAT, on semantic segmentation on ADE20K (3.89%), pose estimation on MSCOCO (9.4%), and reduce the gap with models pre-trained on the object-centric expensive ImageNet-1K. We outperform even when the representations are frozen, and when downstream training data is limited to few-shot. Larger dataset (MIMIC-3M) significantly improves performance, which is promising since our curation method can arbitrarily scale to produce even larger datasets. MIMIC code, dataset, and pretrained models are open-sourced at https://github.com/RAIVNLab/MIMIC.

arXiv:2306.13865

IERL: Interpretable Ensemble Representation Learning -- Combining CrowdSourced Knowledge and Distributed Semantic Representations

Large Language Models (LLMs) encode meanings of words in the form of distributed semantics. Distributed semantics capture common statistical patterns among language tokens (words, phrases, and sentences) from large amounts of data. LLMs perform exceedingly well across General Language Understanding Evaluation (GLUE) tasks designed to test a model's understanding of the meanings of the input tokens. However, recent studies have shown that LLMs tend to generate unintended, inconsistent, or wrong texts as outputs when processing inputs that were seen rarely during training, or inputs that are associated with diverse contexts (e.g., well-known hallucination phenomenon in language generation tasks). Crowdsourced and expert-curated knowledge graphs such as ConceptNet are designed to capture the meaning of words from a compact set of well-defined contexts. Thus LLMs may benefit from leveraging such knowledge contexts to reduce inconsistencies in outputs. We propose a novel ensemble learning method, Interpretable Ensemble Representation Learning (IERL), that systematically combines LLM and crowdsourced knowledge representations of input tokens. IERL has the distinct advantage of being interpretable by design (when was the LLM context used vs. when was the knowledge context used?) over state-of-the-art (SOTA) methods, allowing scrutiny of the inputs in conjunction with the parameters of the model, facilitating the analysis of models' inconsistent or irrelevant outputs. Although IERL is agnostic to the choice of LLM and crowdsourced knowledge, we demonstrate our approach using BERT and ConceptNet. We report improved or competitive results with IERL across GLUE tasks over current SOTA methods and significantly enhanced model interpretability.

arXiv:2306.14080

Leveraging Brain Modularity Prior for Interpretable Representation Learning of fMRI

Resting-state functional magnetic resonance imaging (rs-fMRI) can reflect spontaneous neural activities in brain and is widely used for brain disorder analysis.Previous studies propose to extract fMRI representations through diverse machine/deep learning methods for subsequent analysis. But the learned features typically lack biological interpretability, which limits their clinical utility. From the view of graph theory, the brain exhibits a remarkable modular structure in spontaneous brain functional networks, with each module comprised of functionally interconnected brain regions-of-interest (ROIs). However, most existing learning-based methods for fMRI analysis fail to adequately utilize such brain modularity prior. In this paper, we propose a Brain Modularity-constrained dynamic Representation learning (BMR) framework for interpretable fMRI analysis, consisting of three major components: (1) dynamic graph construction, (2) dynamic graph learning via a novel modularity-constrained graph neural network(MGNN), (3) prediction and biomarker detection for interpretable fMRI analysis. Especially, three core neurocognitive modules (i.e., salience network, central executive network, and default mode network) are explicitly incorporated into the MGNN, encouraging the nodes/ROIs within the same module to share similar representations. To further enhance discriminative ability of learned features, we also encourage the MGNN to preserve the network topology of input graphs via a graph topology reconstruction constraint. Experimental results on 534 subjects with rs-fMRI scans from two datasets validate the effectiveness of the proposed method. The identified discriminative brain ROIs and functional connectivities can be regarded as potential fMRI biomarkers to aid in clinical diagnosis.

arXiv:2306.13926

Graph Neural Networks Provably Benefit from Structural Information: A Feature Learning Perspective

Graph neural networks (GNNs) have pioneered advancements in graph representation learning, exhibiting superior feature learning and performance over multilayer perceptrons (MLPs) when handling graph inputs. However, understanding the feature learning aspect of GNNs is still in its initial stage. This study aims to bridge this gap by investigating the role of graph convolution within the context of feature learning theory in neural networks using gradient descent training. We provide a distinct characterization of signal learning and noise memorization in two-layer graph convolutional networks (GCNs), contrasting them with two-layer convolutional neural networks (CNNs). Our findings reveal that graph convolution significantly augments the benign overfitting regime over the counterpart CNNs, where signal learning surpasses noise memorization, by approximately factor $\sqrt{D}^{q-2}$, with $D$ denoting a node's expected degree and $q$ being the power of the ReLU activation function where $q > 2$. These findings highlight a substantial discrepancy between GNNs and MLPs in terms of feature learning and generalization capacity after gradient descent training, a conclusion further substantiated by our empirical simulations.

arXiv:2306.14861공식 코드 · Papers with Code

Leveraging Task Structures for Improved Identifiability in Neural Network Representations

This work extends the theory of identifiability in supervised learning by considering the consequences of having access to a distribution of tasks. In such cases, we show that linear identifiability is achievable in the general multi-task regression setting. Furthermore, we show that the existence of a task distribution which defines a conditional prior over latent factors reduces the equivalence class for identifiability to permutations and scaling of the true latent factors, a stronger and more useful result than linear identifiability. Crucially, when we further assume a causal structure over these tasks, our approach enables simple maximum marginal likelihood optimization, and suggests potential downstream applications to causal representation learning. Empirically, we find that this straightforward optimization procedure enables our model to outperform more general unsupervised models in recovering canonical representations for both synthetic data and real-world molecular data.

arXiv:2306.14182

Switch-BERT: Learning to Model Multimodal Interactions by Switching Attention and Input

The ability to model intra-modal and inter-modal interactions is fundamental in multimodal machine learning. The current state-of-the-art models usually adopt deep learning models with fixed structures. They can achieve exceptional performances on specific tasks, but face a particularly challenging problem of modality mismatch because of diversity of input modalities and their fixed structures. In this paper, we present \textbf{Switch-BERT} for joint vision and language representation learning to address this problem. Switch-BERT extends BERT architecture by introducing learnable layer-wise and cross-layer interactions. It learns to optimize attention from a set of attention modes representing these interactions. One specific property of the model is that it learns to attend outputs from various depths, therefore mitigates the modality mismatch problem. We present extensive experiments on visual question answering, image-text retrieval and referring expression comprehension experiments. Results confirm that, whereas alternative architectures including ViLBERT and UNITER may excel in particular tasks, Switch-BERT can consistently achieve better or comparable performances than the current state-of-the-art models in these tasks. Ablation studies indicate that the proposed model achieves superior performances due to its ability in learning task-specific multimodal interactions.

arXiv:2306.14701

Hard Sample Mining Enabled Supervised Contrastive Feature Learning for Wind Turbine Pitch System Fault Diagnosis

The efficient utilization of wind power by wind turbines relies on the ability of their pitch systems to adjust blade pitch angles in response to varying wind speeds. However, the presence of multiple health conditions in the pitch system due to the long-term wear and tear poses challenges in accurately classifying them, thus increasing the maintenance cost of wind turbines or even damaging them. This paper proposes a novel method based on hard sample mining-enabled supervised contrastive learning (HSMSCL) to address this problem. The proposed method employs cosine similarity to identify hard samples and subsequently, leverages supervised contrastive learning to learn more discriminative representations by constructing hard sample pairs. Furthermore, the hard sample mining framework in the proposed method also constructs hard samples with learned representations to make the training process of the multilayer perceptron (MLP) more challenging and make it a more effective classifier. The proposed approach progressively improves the fault diagnosis model by introducing hard samples in the SCL and MLP phases, thus enhancing its performance in complex multi-class fault diagnosis tasks. To evaluate the effectiveness of the proposed method, two real datasets comprising wind turbine pitch system cog belt fracture data are utilized. The fault diagnosis performance of the proposed method is compared against existing methods, and the results demonstrate its superior performance. The proposed approach exhibits significant improvements in fault diagnosis performance, providing promising prospects for enhancing the reliability and efficiency of wind turbine pitch system fault diagnosis.

arXiv:2306.14293

Multi-Scale Cross Contrastive Learning for Semi-Supervised Medical Image Segmentation

Semi-supervised learning has demonstrated great potential in medical image segmentation by utilizing knowledge from unlabeled data. However, most existing approaches do not explicitly capture high-level semantic relations between distant regions, which limits their performance. In this paper, we focus on representation learning for semi-supervised learning, by developing a novel Multi-Scale Cross Supervised Contrastive Learning (MCSC) framework, to segment structures in medical images. We jointly train CNN and Transformer models, regularising their features to be semantically consistent across different scales. Our approach contrasts multi-scale features based on ground-truth and cross-predicted labels, in order to extract robust feature representations that reflect intra- and inter-slice relationships across the whole dataset. To tackle class imbalance, we take into account the prevalence of each class to guide contrastive learning and ensure that features adequately capture infrequent classes. Extensive experiments on two multi-structure medical segmentation datasets demonstrate the effectiveness of MCSC. It not only outperforms state-of-the-art semi-supervised methods by more than 3.0% in Dice, but also greatly reduces the performance gap with fully supervised methods.