PyTorchKR Papers

최신 논문

총 4,526편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2210.08388

RoS-KD: A Robust Stochastic Knowledge Distillation Approach for Noisy Medical Imaging

AI-powered Medical Imaging has recently achieved enormous attention due to its ability to provide fast-paced healthcare diagnoses. However, it usually suffers from a lack of high-quality datasets due to high annotation cost, inter-observer variability, human annotator error, and errors in computer-generated labels. Deep learning models trained on noisy labelled datasets are sensitive to the noise type and lead to less generalization on the unseen samples. To address this challenge, we propose a Robust Stochastic Knowledge Distillation (RoS-KD) framework which mimics the notion of learning a topic from multiple sources to ensure deterrence in learning noisy information. More specifically, RoS-KD learns a smooth, well-informed, and robust student manifold by distilling knowledge from multiple teachers trained on overlapping subsets of training data. Our extensive experiments on popular medical imaging classification tasks (cardiopulmonary disease and lesion classification) using real-world datasets, show the performance benefit of RoS-KD, its ability to distill knowledge from many popular large networks (ResNet-50, DenseNet-121, MobileNet-V2) in a comparatively small network, and its robustness to adversarial attacks (PGD, FSGM). More specifically, RoS-KD achieves >2% and >4% improvement on F1-score for lesion classification and cardiopulmonary disease classification tasks, respectively, when the underlying student is ResNet-18 against recent competitive knowledge distillation baseline. Additionally, on cardiopulmonary disease classification task, RoS-KD outperforms most of the SOTA baselines by ~1% gain in AUC score.

arXiv:2210.09022공식 코드 · Papers with Code

Distilling Object Detectors With Global Knowledge

Knowledge distillation learns a lightweight student model that mimics a cumbersome teacher. Existing methods regard the knowledge as the feature of each instance or their relations, which is the instance-level knowledge only from the teacher model, i.e., the local knowledge. However, the empirical studies show that the local knowledge is much noisy in object detection tasks, especially on the blurred, occluded, or small instances. Thus, a more intrinsic approach is to measure the representations of instances w.r.t. a group of common basis vectors in the two feature spaces of the teacher and the student detectors, i.e., global knowledge. Then, the distilling algorithm can be applied as space alignment. To this end, a novel prototype generation module (PGM) is proposed to find the common basis vectors, dubbed prototypes, in the two feature spaces. Then, a robust distilling module (RDM) is applied to construct the global knowledge based on the prototypes and filtrate noisy global and local knowledge by measuring the discrepancy of the representations in two feature spaces. Experiments with Faster-RCNN and RetinaNet on PASCAL and COCO datasets show that our method achieves the best performance for distilling object detectors with various backbones, which even surpasses the performance of the teacher model. We also show that the existing methods can be easily combined with global knowledge and obtain further improvement. Code is available: https://github.com/hikvision-research/DAVAR-Lab-ML.

arXiv:2210.08464

Federated Learning with Privacy-Preserving Ensemble Attention Distillation

Federated Learning (FL) is a machine learning paradigm where many local nodes collaboratively train a central model while keeping the training data decentralized. This is particularly relevant for clinical applications since patient data are usually not allowed to be transferred out of medical facilities, leading to the need for FL. Existing FL methods typically share model parameters or employ co-distillation to address the issue of unbalanced data distribution. However, they also require numerous rounds of synchronized communication and, more importantly, suffer from a privacy leakage risk. We propose a privacy-preserving FL framework leveraging unlabeled public data for one-way offline knowledge distillation in this work. The central model is learned from local knowledge via ensemble attention distillation. Our technique uses decentralized and heterogeneous local data like existing FL approaches, but more importantly, it significantly reduces the risk of privacy leakage. We demonstrate that our method achieves very competitive performance with more robust privacy preservation based on extensive experiments on image classification, segmentation, and reconstruction tasks.

arXiv:2210.08086공식 코드 · Papers with Code

Knowledge Distillation approach towards Melanoma Detection

Melanoma is regarded as the most threatening among all skin cancers. There is a pressing need to build systems which can aid in the early detection of melanoma and enable timely treatment to patients. Recent methods are geared towards machine learning based systems where the task is posed as image recognition, tag dermoscopic images of skin lesions as melanoma or non-melanoma. Even though these methods show promising results in terms of accuracy, they are computationally quite expensive to train, that questions the ability of these models to be deployable in a clinical setting or memory constraint devices. To address this issue, we focus on building simple and performant models having few layers, less than ten compared to hundreds. As well as with fewer learnable parameters, 0.26 million (M) compared to 42.5M using knowledge distillation with the goal to detect melanoma from dermoscopic images. First, we train a teacher model using a ResNet-50 to detect melanoma. Using the teacher model, we train the student model known as Distilled Student Network (DSNet) which has around 0.26M parameters using knowledge distillation achieving an accuracy of 91.7%. We compare against ImageNet pre-trained models such MobileNet, VGG-16, Inception-V3, EfficientNet-B0, ResNet-50 and ResNet-101. We find that our approach works well in terms of inference runtime compared to other pre-trained models, 2.57 seconds compared to 14.55 seconds. We find that DSNet (0.26M parameters), which is 15 times smaller, consistently performs better than EfficientNet-B0 (4M parameters) in both melanoma and non-melanoma detection across Precision, Recall and F1 scores

arXiv:2210.07978공식 코드 · Papers with Code

Improving generalizability of distilled self-supervised speech processing models under distorted settings

Self-supervised learned (SSL) speech pre-trained models perform well across various speech processing tasks. Distilled versions of SSL models have been developed to match the needs of on-device speech applications. Though having similar performance as original SSL models, distilled counterparts suffer from performance degradation even more than their original versions in distorted environments. This paper proposes to apply Cross-Distortion Mapping and Domain Adversarial Training to SSL models during knowledge distillation to alleviate the performance gap caused by the domain mismatch problem. Results show consistent performance improvements under both in- and out-of-domain distorted setups for different downstream tasks while keeping efficient model size.

arXiv:2210.07795공식 코드 · Papers with Code

EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning

Pre-trained vision-language models (VLMs) have achieved impressive results in a range of vision-language tasks. However, popular VLMs usually consist of hundreds of millions of parameters which brings challenges for fine-tuning and deployment in real-world applications due to space, memory, and latency constraints. In this work, we introduce a distilling then pruning framework to compress large vision-language models into smaller, faster, and more accurate ones. We first shrink the size of a pre-trained large VLM and apply knowledge distillation in the vision-language pre-training stage to obtain a task-agnostic compact VLM. Then we propose a modal-adaptive pruning algorithm to automatically infer the importance of vision and language modalities for different downstream tasks and adaptively remove redundant structures and neurons in different encoders with controllable target sparsity. We apply our framework to train EfficientVLM, a fast and accurate vision-language model consisting of 6 vision layers, 3 text layers, and 3 cross-modal fusion layers, accounting for only 93 million parameters in total, which is 44.3% of the teacher model. EfficientVLM retains 98.4% performance of the teacher model and accelerates its inference speed by 2.2x. EfficientVLM achieves a large absolute improvement over previous SoTA efficient VLMs of similar sizes by a large margin on various vision-language tasks, including VQAv2 (+4.9%), NLVR2 (+5.6%), ITR (R@1 on TR +17.2%, on IR + 15.6% ) and COCO caption generation (CIDEr +6.5), demonstrating a large potential on training lightweight VLMs.

arXiv:2210.07686공식 코드 · Papers with Code

Learning Generalizable Models for Vehicle Routing Problems via Knowledge Distillation

Recent neural methods for vehicle routing problems always train and test the deep models on the same instance distribution (i.e., uniform). To tackle the consequent cross-distribution generalization concerns, we bring the knowledge distillation to this field and propose an Adaptive Multi-Distribution Knowledge Distillation (AMDKD) scheme for learning more generalizable deep models. Particularly, our AMDKD leverages various knowledge from multiple teachers trained on exemplar distributions to yield a light-weight yet generalist student model. Meanwhile, we equip AMDKD with an adaptive strategy that allows the student to concentrate on difficult distributions, so as to absorb hard-to-master knowledge more effectively. Extensive experimental results show that, compared with the baseline neural methods, our AMDKD is able to achieve competitive results on both unseen in-distribution and out-of-distribution instances, which are either randomly synthesized or adopted from benchmark datasets (i.e., TSPLIB and CVRPLIB). Notably, our AMDKD is generic, and consumes less computational resources for inference.

arXiv:2210.07135

You Can Have Your Data and Balance It Too: Towards Balanced and Efficient Multilingual Models

Multilingual models have been widely used for cross-lingual transfer to low-resource languages. However, the performance on these languages is hindered by their underrepresentation in the pretraining data. To alleviate this problem, we propose a novel multilingual training technique based on teacher-student knowledge distillation. In this setting, we utilize monolingual teacher models optimized for their language. We use those teachers along with balanced (sub-sampled) data to distill the teachers' knowledge into a single multilingual student. Our method outperforms standard training methods in low-resource languages and retrains performance on high-resource languages while using the same amount of data. If applied widely, our approach can increase the representation of low-resource languages in NLP systems.

arXiv:2210.06980

Probabilistic Integration of Object Level Annotations in Chest X-ray Classification

Medical image datasets and their annotations are not growing as fast as their equivalents in the general domain. This makes translation from the newest, more data-intensive methods that have made a large impact on the vision field increasingly more difficult and less efficient. In this paper, we propose a new probabilistic latent variable model for disease classification in chest X-ray images. Specifically we consider chest X-ray datasets that contain global disease labels, and for a smaller subset contain object level expert annotations in the form of eye gaze patterns and disease bounding boxes. We propose a two-stage optimization algorithm which is able to handle these different label granularities through a single training pipeline in a two-stage manner. In our pipeline global dataset features are learned in the lower level layers of the model. The specific details and nuances in the fine-grained expert object-level annotations are learned in the final layers of the model using a knowledge distillation method inspired by conditional variational inference. Subsequently, model weights are frozen to guide this learning process and prevent overfitting on the smaller richly annotated data subsets. The proposed method yields consistent classification improvement across different backbones on the common benchmark datasets Chest X-ray14 and MIMIC-CXR. This shows how two-stage learning of labels from coarse to fine-grained, in particular with object level annotations, is an effective method for more optimal annotation usage.

arXiv:2210.06297공식 코드 · Papers with Code

Multimodality Multi-Lead ECG Arrhythmia Classification using Self-Supervised Learning

Electrocardiogram (ECG) signal is one of the most effective sources of information mainly employed for the diagnosis and prediction of cardiovascular diseases (CVDs) connected with the abnormalities in heart rhythm. Clearly, single modality ECG (i.e. time series) cannot convey its complete characteristics, thus, exploiting both time and time-frequency modalities in the form of time-series data and spectrogram is needed. Leveraging the cutting-edge self-supervised learning (SSL) technique on unlabeled data, we propose SSL-based multimodality ECG classification. Our proposed network follows SSL learning paradigm and consists of two modules corresponding to pre-stream task, and down-stream task, respectively. In the SSL-pre-stream task, we utilize self-knowledge distillation (KD) techniques with no labeled data, on various transformations and in both time and frequency domains. In the down-stream task, which is trained on labeled data, we propose a gate fusion mechanism to fuse information from multimodality.To evaluate the effectiveness of our approach, ten-fold cross validation on the 12-lead PhysioNet 2020 dataset has been conducted.

arXiv:2210.06020공식 코드 · Papers with Code

Integrating Translation Memories into Non-Autoregressive Machine Translation

Non-autoregressive machine translation (NAT) has recently made great progress. However, most works to date have focused on standard translation tasks, even though some edit-based NAT models, such as the Levenshtein Transformer (LevT), seem well suited to translate with a Translation Memory (TM). This is the scenario considered here. We first analyze the vanilla LevT model and explain why it does not do well in this setting. We then propose a new variant, TM-LevT, and show how to effectively train this model. By modifying the data presentation and introducing an extra deletion operation, we obtain performance that are on par with an autoregressive approach, while reducing the decoding load. We also show that incorporating TMs during training dispenses to use knowledge distillation, a well-known trick used to mitigate the multimodality issue.

arXiv:2210.05920

Boosting Graph Neural Networks via Adaptive Knowledge Distillation

Graph neural networks (GNNs) have shown remarkable performance on diverse graph mining tasks. Although different GNNs can be unified as the same message passing framework, they learn complementary knowledge from the same graph. Knowledge distillation (KD) is developed to combine the diverse knowledge from multiple models. It transfers knowledge from high-capacity teachers to a lightweight student. However, to avoid oversmoothing, GNNs are often shallow, which deviates from the setting of KD. In this context, we revisit KD by separating its benefits from model compression and emphasizing its power of transferring knowledge. To this end, we need to tackle two challenges: how to transfer knowledge from compact teachers to a student with the same capacity; and, how to exploit student GNN's own strength to learn knowledge. In this paper, we propose a novel adaptive KD framework, called BGNN, which sequentially transfers knowledge from multiple GNNs into a student GNN. We also introduce an adaptive temperature module and a weight boosting module. These modules guide the student to the appropriate knowledge for effective learning. Extensive experiments have demonstrated the effectiveness of BGNN. In particular, we achieve up to 3.05% improvement for node classification and 6.35% improvement for graph classification over vanilla GNNs.

arXiv:2210.05801

Linkless Link Prediction via Relational Distillation

Graph Neural Networks (GNNs) have shown exceptional performance in the task of link prediction. Despite their effectiveness, the high latency brought by non-trivial neighborhood data dependency limits GNNs in practical deployments. Conversely, the known efficient MLPs are much less effective than GNNs due to the lack of relational knowledge. In this work, to combine the advantages of GNNs and MLPs, we start with exploring direct knowledge distillation (KD) methods for link prediction, i.e., predicted logit-based matching and node representation-based matching. Upon observing direct KD analogs do not perform well for link prediction, we propose a relational KD framework, Linkless Link Prediction (LLP), to distill knowledge for link prediction with MLPs. Unlike simple KD methods that match independent link logits or node representations, LLP distills relational knowledge that is centered around each (anchor) node to the student MLP. Specifically, we propose rank-based matching and distribution-based matching strategies that complement each other. Extensive experiments demonstrate that LLP boosts the link prediction performance of MLPs with significant margins, and even outperforms the teacher GNNs on 7 out of 8 benchmarks. LLP also achieves a 70.68x speedup in link prediction inference compared to GNNs on the large-scale OGB dataset.

arXiv:2210.05793

Comparison of Soft and Hard Target RNN-T Distillation for Large-scale ASR

Knowledge distillation is an effective machine learning technique to transfer knowledge from a teacher model to a smaller student model, especially with unlabeled data. In this paper, we focus on knowledge distillation for the RNN-T model, which is widely used in state-of-the-art (SoTA) automatic speech recognition (ASR). Specifically, we compared using soft and hard target distillation to train large-scaleRNN-T models on the LibriSpeech/LibriLight public dataset (60k hours) and our in-house data (600k hours). We found that hard tar-gets are more effective when the teacher and student have different architecture, such as large teacher and small streaming student. On the other hand, soft target distillation works better in self-training scenario like iterative large teacher training. For a large model with0.6B weights, we achieve a new SoTA word error rate (WER) on LibriSpeech (8% relative improvement on dev-other) using Noisy Student Training with soft target distillation. It also allows our production teacher to adapt new data domain continuously.

arXiv:2210.05941

Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Class-incremental semantic segmentation (CISS) labels each pixel of an image with a corresponding object/stuff class continually. To this end, it is crucial to learn novel classes incrementally without forgetting previously learned knowledge. Current CISS methods typically use a knowledge distillation (KD) technique for preserving classifier logits, or freeze a feature extractor, to avoid the forgetting problem. The strong constraints, however, prevent learning discriminative features for novel classes. We introduce a CISS framework that alleviates the forgetting problem and facilitates learning novel classes effectively. We have found that a logit can be decomposed into two terms. They quantify how likely an input belongs to a particular class or not, providing a clue for a reasoning process of a model. The KD technique, in this context, preserves the sum of two terms (i.e., a class logit), suggesting that each could be changed and thus the KD does not imitate the reasoning process. To impose constraints on each term explicitly, we propose a new decomposed knowledge distillation (DKD) technique, improving the rigidity of a model and addressing the forgetting problem more effectively. We also introduce a novel initialization method to train new classifiers for novel classes. In CISS, the number of negative training samples for novel classes is not sufficient to discriminate old classes. To mitigate this, we propose to transfer knowledge of negatives to the classifiers successively using an auxiliary classifier, boosting the performance significantly. Experimental results on standard CISS benchmarks demonstrate the effectiveness of our framework.

arXiv:2210.05832

SaiT: Sparse Vision Transformers through Adaptive Token Pruning

While vision transformers have achieved impressive results, effectively and efficiently accelerating these models can further boost performances. In this work, we propose a dense/sparse training framework to obtain a unified model, enabling weight sharing across various token densities. Thus one model offers a range of accuracy and throughput tradeoffs for different applications. Besides, we introduce adaptive token pruning to optimize the patch token sparsity based on the input image. In addition, we investigate knowledge distillation to enhance token selection capability in early transformer modules. Sparse adaptive image Transformer (SaiT) offers varying levels of model acceleration by merely changing the token sparsity on the fly. Specifically, SaiT reduces the computation complexity (FLOPs) by 39% - 43% and increases the throughput by 67% - 91% with less than 0.5% accuracy loss for various vision transformer models. Meanwhile, the same model also provides the zero accuracy drop option by skipping the sparsification step. SaiT achieves better accuracy and computation tradeoffs than state-of-the-art transformer and convolutional models.

arXiv:2210.06458공식 코드 · Papers with Code

Efficient Knowledge Distillation from Model Checkpoints

Knowledge distillation is an effective approach to learn compact models (students) with the supervision of large and strong models (teachers). As empirically there exists a strong correlation between the performance of teacher and student models, it is commonly believed that a high performing teacher is preferred. Consequently, practitioners tend to use a well trained network or an ensemble of them as the teacher. In this paper, we make an intriguing observation that an intermediate model, i.e., a checkpoint in the middle of the training procedure, often serves as a better teacher compared to the fully converged model, although the former has much lower accuracy. More surprisingly, a weak snapshot ensemble of several intermediate models from a same training trajectory can outperform a strong ensemble of independently trained and fully converged models, when they are used as teachers. We show that this phenomenon can be partially explained by the information bottleneck principle: the feature representations of intermediate models can have higher mutual information regarding the input, and thus contain more "dark knowledge" for effective distillation. We further propose an optimal intermediate teacher selection algorithm based on maximizing the total task-related mutual information. Experiments verify its effectiveness and applicability.

arXiv:2210.05508공식 코드 · Papers with Code

Detect, Distill and Update: Learned DB Systems Facing Out of Distribution Data

Machine Learning (ML) is changing DBs as many DB components are being replaced by ML models. One open problem in this setting is how to update such ML models in the presence of data updates. We start this investigation focusing on data insertions (dominating updates in analytical DBs). We study how to update neural network (NN) models when new data follows a different distribution (a.k.a. it is "out-of-distribution" -- OOD), rendering previously-trained NNs inaccurate. A requirement in our problem setting is that learned DB components should ensure high accuracy for tasks on old and new data (e.g., for approximate query processing (AQP), cardinality estimation (CE), synthetic data generation (DG), etc.). This paper proposes a novel updatability framework (DDUp). DDUp can provide updatability for different learned DB system components, even based on different NNs, without the high costs to retrain the NNs from scratch. DDUp entails two components: First, a novel, efficient, and principled statistical-testing approach to detect OOD data. Second, a novel model updating approach, grounded on the principles of transfer learning with knowledge distillation, to update learned models efficiently, while still ensuring high accuracy. We develop and showcase DDUp's applicability for three different learned DB components, AQP, CE, and DG, each employing a different type of NN. Detailed experimental evaluation using real and benchmark datasets for AQP, CE, and DG detail DDUp's performance advantages.

arXiv:2210.05638공식 코드 · Papers with Code

APSNet: Attention Based Point Cloud Sampling

Processing large point clouds is a challenging task. Therefore, the data is often downsampled to a smaller size such that it can be stored, transmitted and processed more efficiently without incurring significant performance degradation. Traditional task-agnostic sampling methods, such as farthest point sampling (FPS), do not consider downstream tasks when sampling point clouds, and thus non-informative points to the tasks are often sampled. This paper explores a task-oriented sampling for 3D point clouds, and aims to sample a subset of points that are tailored specifically to a downstream task of interest. Similar to FPS, we assume that point to be sampled next should depend heavily on the points that have already been sampled. We thus formulate point cloud sampling as a sequential generation process, and develop an attention-based point cloud sampling network (APSNet) to tackle this problem. At each time step, APSNet attends to all the points in a cloud by utilizing the history of previously sampled points, and samples the most informative one. Both supervised learning and knowledge distillation-based self-supervised learning of APSNet are proposed. Moreover, joint training of APSNet over multiple sample sizes is investigated, leading to a single APSNet that can generate arbitrary length of samples with prominent performances. Extensive experiments demonstrate the superior performance of APSNet against state-of-the-arts in various downstream tasks, including 3D point cloud classification, reconstruction, and registration.

arXiv:2210.05391공식 코드 · Papers with Code

PP-StructureV2: A Stronger Document Analysis System

A large amount of document data exists in unstructured form such as raw images without any text information. Designing a practical document image analysis system is a meaningful but challenging task. In previous work, we proposed an intelligent document analysis system PP-Structure. In order to further upgrade the function and performance of PP-Structure, we propose PP-StructureV2 in this work, which contains two subsystems: Layout Information Extraction and Key Information Extraction. Firstly, we integrate Image Direction Correction module and Layout Restoration module to enhance the functionality of the system. Secondly, 8 practical strategies are utilized in PP-StructureV2 for better performance. For Layout Analysis model, we introduce ultra light-weight detector PP-PicoDet and knowledge distillation algorithm FGD for model lightweighting, which increased the inference speed by 11 times with comparable mAP. For Table Recognition model, we utilize PP-LCNet, CSP-PAN and SLAHead to optimize the backbone module, feature fusion module and decoding module, respectively, which improved the table structure accuracy by 6\% with comparable inference speed. For Key Information Extraction model, we introduce VI-LayoutXLM which is a visual-feature independent LayoutXLM architecture, TB-YX sorting algorithm and U-DML knowledge distillation algorithm, which brought 2.8\% and 9.1\% improvement respectively on the Hmean of Semantic Entity Recognition and Relation Extraction tasks. All the above mentioned models and code are open-sourced in the GitHub repository PaddleOCR.