PyTorchKR Papers

최신 논문

총 4,806편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2408.02859공식 코드 · Papers with Code

Multistain Pretraining for Slide Representation Learning in Pathology

Developing self-supervised learning (SSL) models that can learn universal and transferable representations of H&E gigapixel whole-slide images (WSIs) is becoming increasingly valuable in computational pathology. These models hold the potential to advance critical tasks such as few-shot classification, slide retrieval, and patient stratification. Existing approaches for slide representation learning extend the principles of SSL from small images (e.g., 224 x 224 patches) to entire slides, usually by aligning two different augmentations (or views) of the slide. Yet the resulting representation remains constrained by the limited clinical and biological diversity of the views. Instead, we postulate that slides stained with multiple markers, such as immunohistochemistry, can be used as different views to form a rich task-agnostic training signal. To this end, we introduce Madeleine, a multimodal pretraining strategy for slide representation learning. Madeleine is trained with a dual global-local cross-stain alignment objective on large cohorts of breast cancer samples (N=4,211 WSIs across five stains) and kidney transplant samples (N=12,070 WSIs across four stains). We demonstrate the quality of slide representations learned by Madeleine on various downstream evaluations, ranging from morphological and molecular classification to prognostic prediction, comprising 21 tasks using 7,299 WSIs from multiple medical centers. Code is available at https://github.com/mahmoodlab/MADELEINE.

arXiv:2408.02945

Self-Supervised Learning for Multi-Channel Neural Transducer

Self-supervised learning, such as with the wav2vec 2.0 framework significantly improves the accuracy of end-to-end automatic speech recognition (ASR). Wav2vec 2.0 has been applied to single-channel end-to-end ASR models. In this work, we explored a self-supervised learning method for a multi-channel end-to-end ASR model based on the wav2vec 2.0 framework. As the multi-channel end-to-end ASR model, we focused on a multi-channel neural transducer. In pre-training, we compared three different methods for feature quantization to train a multi-channel conformer audio encoder: joint quantization, feature-wise quantization and channel-wise quantization. In fine-tuning, we trained the multi-channel conformer-transducer. All experiments were conducted using the far-field in-house and CHiME-4 datasets. The results of the experiments showed that feature-wise quantization was the most effective among the methods. We observed a 66% relative reduction in character error rate compared with the model without any pre-training for the far-field in-house dataset.

arXiv:2408.02091

Past Movements-Guided Motion Representation Learning for Human Motion Prediction

Human motion prediction based on 3D skeleton is a significant challenge in computer vision, primarily focusing on the effective representation of motion. In this paper, we propose a self-supervised learning framework designed to enhance motion representation. This framework consists of two stages: first, the network is pretrained through the self-reconstruction of past sequences, and the guided reconstruction of future sequences based on past movements. We design a velocity-based mask strategy to focus on the joints with large-scale moving. Subsequently, the pretrained network undergoes finetuning for specific tasks. Self-reconstruction, guided by patterns of past motion, substantially improves the model's ability to represent the spatiotemporal relationships among joints but also captures the latent relationships between past and future sequences. This capability is crucial for motion prediction tasks that solely depend on historical motion data. By employing this straightforward yet effective training paradigm, our method outperforms existing \textit{state-of-the-art} methods, reducing the average prediction errors by 8.8\% across Human3.6M, 3DPW, and AMASS datasets. The code is available at https://github.com/JunyuShi02/PMG-MRL.

arXiv:2408.01565

Embodiment: Self-Supervised Depth Estimation Based on Camera Models

Depth estimation is a critical topic for robotics and vision-related tasks. In monocular depth estimation, in comparison with supervised learning that requires expensive ground truth labeling, self-supervised methods possess great potential due to no labeling cost. However, self-supervised learning still has a large gap with supervised learning in 3D reconstruction and depth estimation performance. Meanwhile, scaling is also a major issue for monocular unsupervised depth estimation, which commonly still needs ground truth scale from GPS, LiDAR, or existing maps to correct. In the era of deep learning, existing methods primarily rely on exploring image relationships to train unsupervised neural networks, while the physical properties of the camera itself such as intrinsics and extrinsics are often overlooked. These physical properties are not just mathematical parameters; they are embodiments of the camera's interaction with the physical world. By embedding these physical properties into the deep learning model, we can calculate depth priors for ground regions and regions connected to the ground based on physical principles, providing free supervision signals without the need for additional sensors. This approach is not only easy to implement but also enhances the effects of all unsupervised methods by embedding the camera's physical properties into the model, thereby achieving an embodied understanding of the real world.

arXiv:2408.02514공식 코드 · Papers with Code

Stem-JEPA: A Joint-Embedding Predictive Architecture for Musical Stem Compatibility Estimation

This paper explores the automated process of determining stem compatibility by identifying audio recordings of single instruments that blend well with a given musical context. To tackle this challenge, we present Stem-JEPA, a novel Joint-Embedding Predictive Architecture (JEPA) trained on a multi-track dataset using a self-supervised learning approach. Our model comprises two networks: an encoder and a predictor, which are jointly trained to predict the embeddings of compatible stems from the embeddings of a given context, typically a mix of several instruments. Training a model in this manner allows its use in estimating stem compatibility - retrieving, aligning, or generating a stem to match a given mix - or for downstream tasks such as genre or key estimation, as the training paradigm requires the model to learn information related to timbre, harmony, and rhythm. We evaluate our model's performance on a retrieval task on the MUSDB18 dataset, testing its ability to find the missing stem from a mix and through a subjective user study. We also show that the learned embeddings capture temporal alignment information and, finally, evaluate the representations learned by our model on several downstream tasks, highlighting that they effectively capture meaningful musical features.

arXiv:2408.01705

Downstream Transfer Attack: Adversarial Attacks on Downstream Models with Pre-trained Vision Transformers

With the advancement of vision transformers (ViTs) and self-supervised learning (SSL) techniques, pre-trained large ViTs have become the new foundation models for computer vision applications. However, studies have shown that, like convolutional neural networks (CNNs), ViTs are also susceptible to adversarial attacks, where subtle perturbations in the input can fool the model into making false predictions. This paper studies the transferability of such an adversarial vulnerability from a pre-trained ViT model to downstream tasks. We focus on \emph{sample-wise} transfer attacks and propose a novel attack method termed \emph{Downstream Transfer Attack (DTA)}. For a given test image, DTA leverages a pre-trained ViT model to craft the adversarial example and then applies the adversarial example to attack a fine-tuned version of the model on a downstream dataset. During the attack, DTA identifies and exploits the most vulnerable layers of the pre-trained model guided by a cosine similarity loss to craft highly transferable attacks. Through extensive experiments with pre-trained ViTs by 3 distinct pre-training methods, 3 fine-tuning schemes, and across 10 diverse downstream datasets, we show that DTA achieves an average attack success rate (ASR) exceeding 90\%, surpassing existing methods by a huge margin. When used with adversarial training, the adversarial examples generated by our DTA can significantly improve the model's robustness to different downstream transfer attacks.

arXiv:2408.01427

Siamese Transformer Networks for Few-shot Image Classification

Humans exhibit remarkable proficiency in visual classification tasks, accurately recognizing and classifying new images with minimal examples. This ability is attributed to their capacity to focus on details and identify common features between previously seen and new images. In contrast, existing few-shot image classification methods often emphasize either global features or local features, with few studies considering the integration of both. To address this limitation, we propose a novel approach based on the Siamese Transformer Network (STN). Our method employs two parallel branch networks utilizing the pre-trained Vision Transformer (ViT) architecture to extract global and local features, respectively. Specifically, we implement the ViT-Small network architecture and initialize the branch networks with pre-trained model parameters obtained through self-supervised learning. We apply the Euclidean distance measure to the global features and the Kullback-Leibler (KL) divergence measure to the local features. To integrate the two metrics, we first employ L2 normalization and then weight the normalized results to obtain the final similarity score. This strategy leverages the advantages of both global and local features while ensuring their complementary benefits. During the training phase, we adopt a meta-learning approach to fine-tune the entire network. Our strategy effectively harnesses the potential of global and local features in few-shot image classification, circumventing the need for complex feature adaptation modules and enhancing the model's generalization ability. Extensive experiments demonstrate that our framework is simple yet effective, achieving superior performance compared to state-of-the-art baselines on four popular few-shot classification benchmarks in both 5-shot and 1-shot scenarios.

arXiv:2408.01167공식 코드 · Papers with Code

Rethinking Pre-Trained Feature Extractor Selection in Multiple Instance Learning for Whole Slide Image Classification

Multiple instance learning (MIL) has become a preferred method for gigapixel whole slide image (WSI) classification without requiring patch-level annotations. Current MIL research primarily relies on embedding-based approaches, which extract patch features using a pre-trained feature extractor and aggregate them for slide-level prediction. Despite the critical role of feature extraction, there is limited guidance on selecting optimal feature extractors to maximize WSI performance. This study addresses this gap by systematically evaluating MIL feature extractors across three dimensions: pre-training dataset, backbone model, and pre-training method. Extensive experiments were conducted on two public WSI datasets (TCGA-NSCLC and Camelyon16) using four state-of-the-art (SOTA) MIL models. Our findings reveal that: 1) selecting a robust self-supervised learning (SSL) method has a greater impact on performance than relying solely on an in-domain pre-training dataset; 2) prioritizing Transformer-based backbones with deeper architectures over CNN-based models; and 3) using larger, more diverse pre-training datasets significantly enhances classification outcomes. We hope that these insights can provide practical guidance for optimizing WSI classification and explain the reasons behind the performance advantages of the current SOTA pathology foundation models. Furthermore, this work may inform the development of more effective pathology foundation models. Our code is publicly available at https://github.com/bryanwong17/MIL-Feature-Extractor-Selection

arXiv:2408.00498

How Effective are Self-Supervised Models for Contact Identification in Videos

The exploration of video content via Self-Supervised Learning (SSL) models has unveiled a dynamic field of study, emphasizing both the complex challenges and unique opportunities inherent in this area. Despite the growing body of research, the ability of SSL models to detect physical contacts in videos remains largely unexplored, particularly the effectiveness of methods such as downstream supervision with linear probing or full fine-tuning. This work aims to bridge this gap by employing eight different convolutional neural networks (CNNs) based video SSL models to identify instances of physical contact within video sequences specifically. The Something-Something v2 (SSv2) and Epic-Kitchen (EK-100) datasets were chosen for evaluating these approaches due to the promising results on UCF101 and HMDB51, coupled with their limited prior assessment on SSv2 and EK-100. Additionally, these datasets feature diverse environments and scenarios, essential for testing the robustness and accuracy of video-based models. This approach not only examines the effectiveness of each model in recognizing physical contacts but also explores the performance in the action recognition downstream task. By doing so, valuable insights into the adaptability of SSL models in interpreting complex, dynamic visual information are contributed.

arXiv:2408.00347공식 코드 · Papers with Code

Advancing Medical Image Segmentation: Morphology-Driven Learning with Diffusion Transformer

Understanding the morphological structure of medical images and precisely segmenting the region of interest or abnormality is an important task that can assist in diagnosis. However, the unique properties of medical imaging make clear segmentation difficult,and the high cost and time-consuming task of labeling leads to a coarse-grained representation of ground truth. Facing with these problems, we propose a novel Diffusion Transformer Segmentation (DTS) model for robust segmentation in the presence of noise. We propose an alternative to the dominant Denoising U-Net encoder through experiments applying a transformer architecture, which captures global dependency through self-attention. Additionally, we propose k-neighbor label smoothing, reverse boundary attention, and self-supervised learning with morphology-driven learning to improve the ability to identify complex structures. Our model, which analyzes the morphological representation of images, shows better results than the previous models in various medical imaging modalities, including CT, MRI, and lesion images.

arXiv:2408.00527

Contrastive Learning with Adaptive Neighborhoods for Brain Age Prediction on 3D Stiffness Maps

In the field of neuroimaging, accurate brain age prediction is pivotal for uncovering the complexities of brain aging and pinpointing early indicators of neurodegenerative conditions. Recent advancements in self-supervised learning, particularly in contrastive learning, have demonstrated greater robustness when dealing with complex datasets. However, current approaches often fall short in generalizing across non-uniformly distributed data, prevalent in medical imaging scenarios. To bridge this gap, we introduce a novel contrastive loss that adapts dynamically during the training process, focusing on the localized neighborhoods of samples. Moreover, we expand beyond traditional structural features by incorporating brain stiffness - a mechanical property previously underexplored yet promising due to its sensitivity to age-related changes. This work presents the first application of self-supervised learning to brain mechanical properties, using compiled stiffness maps from various clinical studies to predict brain age. Our approach, featuring dynamic localized loss, consistently outperforms existing state-of-the-art methods, demonstrating superior performance and paving the way for new directions in brain aging research.

arXiv:2408.00380공식 코드 · Papers with Code

EXAONEPath 1.0 Patch-level Foundation Model for Pathology

Recent advancements in digital pathology have led to the development of numerous foundational models that utilize self-supervised learning on patches extracted from gigapixel whole slide images (WSIs). While this approach leverages vast amounts of unlabeled data, we have discovered a significant issue: features extracted from these self-supervised models tend to cluster by individual WSIs, a phenomenon we term WSI-specific feature collapse. This problem can potentially limit the model's generalization ability and performance on various downstream tasks. To address this issue, we introduce EXAONEPath, a novel foundational model trained on patches that have undergone stain normalization. Stain normalization helps reduce color variability arising from different laboratories and scanners, enabling the model to learn more consistent features. EXAONEPath is trained using 285,153,903 patches extracted from a total of 34,795 WSIs. Our experiments demonstrate that EXAONEPath significantly mitigates the feature collapse problem, indicating that the model has learned more generalized features rather than overfitting to individual WSI characteristics. We compared EXAONEPath with state-of-the-art models across six downstream task datasets, and our results show that EXAONEPath achieves superior performance relative to the number of WSIs used and the model's parameter count. This suggests that the application of stain normalization has substantially improved the model's efficiency and generalization capabilities.

arXiv:2408.00256

Mobility-Aware Federated Self-supervised Learning in Vehicular Network

Federated Learning (FL) is an advanced distributed machine learning approach, that protects the privacy of each vehicle by allowing the model to be trained on multiple devices simultaneously without the need to upload all data to a road side unit (RSU). This enables FL to handle scenarios with sensitive or widely distributed data. However, in these fields, it is well known that the labeling costs can be a significant expense, and models relying on labels are not suitable for these rapidly evolving fields especially in vehicular networks, or mobile internet of things (MIoT), where new data emerges constantly. To handle this issue, the self-supervised learning paves the way for training without labels. Additionally, for vehicles with high velocity, owing to blurred images, simple aggregation not only impacts the accuracy of the aggregated model but also reduces the convergence speed of FL. This paper proposes a FL algorithm based on image blur level to aggregation, called FLSimCo, which does not require labels and serves as a pre-training stage for self-supervised learning in the vehicular environment. Simulation results demonstrate that the proposed algorithm exhibits fast and stable convergence.

arXiv:2407.21066

ELP-Adapters: Parameter Efficient Adapter Tuning for Various Speech Processing Tasks

Self-supervised learning has emerged as a key approach for learning generic representations from speech data. Despite promising results in downstream tasks such as speech recognition, speaker verification, and emotion recognition, a significant number of parameters is required, which makes fine-tuning for each task memory-inefficient. To address this limitation, we introduce ELP-adapter tuning, a novel method for parameter-efficient fine-tuning using three types of adapter, namely encoder adapters (E-adapters), layer adapters (L-adapters), and a prompt adapter (P-adapter). The E-adapters are integrated into transformer-based encoder layers and help to learn fine-grained speech representations that are effective for speech recognition. The L-adapters create paths from each encoder layer to the downstream head and help to extract non-linguistic features from lower encoder layers that are effective for speaker verification and emotion recognition. The P-adapter appends pseudo features to CNN features to further improve effectiveness and efficiency. With these adapters, models can be quickly adapted to various speech processing tasks. Our evaluation across four downstream tasks using five backbone models demonstrated the effectiveness of the proposed method. With the WavLM backbone, its performance was comparable to or better than that of full fine-tuning on all tasks while requiring 90% fewer learnable parameters.

arXiv:2407.20267공식 코드 · Papers with Code

A Large Encoder-Decoder Family of Foundation Models For Chemical Language

Large-scale pre-training methodologies for chemical language models represent a breakthrough in cheminformatics. These methods excel in tasks such as property prediction and molecule generation by learning contextualized representations of input tokens through self-supervised learning on large unlabeled corpora. Typically, this involves pre-training on unlabeled data followed by fine-tuning on specific tasks, reducing dependence on annotated datasets and broadening chemical language representation understanding. This paper introduces a large encoder-decoder chemical foundation models pre-trained on a curated dataset of 91 million SMILES samples sourced from PubChem, which is equivalent to 4 billion of molecular tokens. The proposed foundation model supports different complex tasks, including quantum property prediction, and offer flexibility with two main variants (289M and $8\times289M$). Our experiments across multiple benchmark datasets validate the capacity of the proposed model in providing state-of-the-art results for different tasks. We also provide a preliminary assessment of the compositionality of the embedding space as a prerequisite for the reasoning tasks. We demonstrate that the produced latent space is separable compared to the state-of-the-art with few-shot learning capabilities.

arXiv:2407.20395

Dense Self-Supervised Learning for Medical Image Segmentation

Deep learning has revolutionized medical image segmentation, but it relies heavily on high-quality annotations. The time, cost and expertise required to label images at the pixel-level for each new task has slowed down widespread adoption of the paradigm. We propose Pix2Rep, a self-supervised learning (SSL) approach for few-shot segmentation, that reduces the manual annotation burden by learning powerful pixel-level representations directly from unlabeled images. Pix2Rep is a novel pixel-level loss and pre-training paradigm for contrastive SSL on whole images. It is applied to generic encoder-decoder deep learning backbones (e.g., U-Net). Whereas most SSL methods enforce invariance of the learned image-level representations under intensity and spatial image augmentations, Pix2Rep enforces equivariance of the pixel-level representations. We demonstrate the framework on a task of cardiac MRI segmentation. Results show improved performance compared to existing semi- and self-supervised approaches; and a 5-fold reduction in the annotation burden for equivalent performance versus a fully supervised U-Net baseline. This includes a 30% (resp. 31%) DICE improvement for one-shot segmentation under linear-probing (resp. fine-tuning). Finally, we also integrate the novel Pix2Rep concept with the Barlow Twins non-contrastive SSL, which leads to even better segmentation performance.

arXiv:2407.19692

Fusion Self-supervised Learning for Recommendation

Recommender systems are widely deployed in various web environments, and self-supervised learning (SSL) has recently attracted significant attention in this field. Contrastive learning (CL) stands out as a major SSL paradigm due to its robust ability to generate self-supervised signals. Mainstream graph contrastive learning (GCL)-based methods typically implement CL by creating contrastive views through various data augmentation techniques. Despite these methods are effective, we argue that there still exist several challenges. i) Data augmentation ($e.g.,$ discarding edges or adding noise) necessitates additional graph convolution (GCN) or modeling operations, which are highly time-consuming and potentially harm the embedding quality. ii) Existing CL-based methods use traditional CL objectives to capture self-supervised signals. However, few studies have explored obtaining CL objectives from more perspectives and have attempted to fuse the varying signals from these CL objectives to enhance recommendation performance. To overcome these challenges, we propose a Fusion Self-supervised Learning framework for recommendation. Specifically, instead of facilitating data augmentations, we use high-order information from GCN process to create contrastive views. Additionally, to integrate self-supervised signals from various CL objectives, we propose an advanced CL objective. By ensuring that positive pairs are distanced from negative samples derived from both contrastive views, we effectively fuse self-supervised signals from distinct CL objectives, thereby enhancing the mutual information between positive pairs. Experimental results on three public datasets demonstrate the superior recommendation performance and efficiency of HFGCL compared to the state-of-the-art baselines.

arXiv:2407.19658

Enhancing CTR Prediction through Sequential Recommendation Pre-training: Introducing the SRP4CTR Framework

Understanding user interests is crucial for Click-Through Rate (CTR) prediction tasks. In sequential recommendation, pre-training from user historical behaviors through self-supervised learning can better comprehend user dynamic preferences, presenting the potential for direct integration with CTR tasks. Previous methods have integrated pre-trained models into downstream tasks with the sole purpose of extracting semantic information or well-represented user features, which are then incorporated as new features. However, these approaches tend to ignore the additional inference costs to the downstream tasks, and they do not consider how to transfer the effective information from the pre-trained models for specific estimated items in CTR prediction. In this paper, we propose a Sequential Recommendation Pre-training framework for CTR prediction (SRP4CTR) to tackle the above problems. Initially, we discuss the impact of introducing pre-trained models on inference costs. Subsequently, we introduced a pre-trained method to encode sequence side information concurrently.During the fine-tuning process, we incorporate a cross-attention block to establish a bridge between estimated items and the pre-trained model at a low cost. Moreover, we develop a querying transformer technique to facilitate the knowledge transfer from the pre-trained model to industrial CTR models. Offline and online experiments show that our method outperforms previous baseline models.

arXiv:2407.19889

Self-Supervised Learning for Text Recognition: A Critical Survey

Text Recognition (TR) refers to the research area that focuses on retrieving textual information from images, a topic that has seen significant advancements in the last decade due to the use of Deep Neural Networks (DNN). However, these solutions often necessitate vast amounts of manually labeled or synthetic data. Addressing this challenge, Self-Supervised Learning (SSL) has gained attention by utilizing large datasets of unlabeled data to train DNN, thereby generating meaningful and robust representations. Although SSL was initially overlooked in TR because of its unique characteristics, recent years have witnessed a surge in the development of SSL methods specifically for this field. This rapid development, however, has led to many methods being explored independently, without taking previous efforts in methodology or comparison into account, thereby hindering progress in the field of research. This paper, therefore, seeks to consolidate the use of SSL in the field of TR, offering a critical and comprehensive overview of the current state of the art. We will review and analyze the existing methods, compare their results, and highlight inconsistencies in the current literature. This thorough analysis aims to provide general insights into the field, propose standardizations, identify new research directions, and foster its proper development.

arXiv:2407.18332

Analyzing Speech Unit Selection for Textless Speech-to-Speech Translation

Recent advancements in textless speech-to-speech translation systems have been driven by the adoption of self-supervised learning techniques. Although most state-of-the-art systems adopt a similar architecture to transform source language speech into sequences of discrete representations in the target language, the criteria for selecting these target speech units remains an open question. This work explores the selection process through a study of downstream tasks such as automatic speech recognition, speech synthesis, speaker recognition, and emotion recognition. Interestingly, our findings reveal a discrepancy in the optimization of discrete speech units: units that perform well in resynthesis performance do not necessarily correlate with those that enhance translation efficacy. This discrepancy underscores the nuanced complexity of target feature selection and its impact on the overall performance of speech-to-speech translation systems.