PyTorchKR Papers

최신 논문

총 4,526편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2210.05280공식 코드 · Papers with Code

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

Recently, Cross-Domain Few-Shot Learning (CD-FSL) which aims at addressing the Few-Shot Learning (FSL) problem across different domains has attracted rising attention. The core challenge of CD-FSL lies in the domain gap between the source and novel target datasets. Though many attempts have been made for CD-FSL without any target data during model training, the huge domain gap makes it still hard for existing CD-FSL methods to achieve very satisfactory results. Alternatively, learning CD-FSL models with few labeled target domain data which is more realistic and promising is advocated in previous work~\cite{fu2021meta}. Thus, in this paper, we stick to this setting and technically contribute a novel Multi-Expert Domain Decompositional Network (ME-D2N). Concretely, to solve the data imbalance problem between the source data with sufficient examples and the auxiliary target data with limited examples, we build our model under the umbrella of multi-expert learning. Two teacher models which can be considered to be experts in their corresponding domain are first trained on the source and the auxiliary target sets, respectively. Then, the knowledge distillation technique is introduced to transfer the knowledge from two teachers to a unified student model. Taking a step further, to help our student model learn knowledge from different domain teachers simultaneously, we further present a novel domain decomposition module that learns to decompose the student model into two domain-related sub parts. This is achieved by a novel domain-specific gate that learns to assign each filter to only one specific domain in a learnable way. Extensive experiments demonstrate the effectiveness of our method. Codes and models are available at https://github.com/lovelyqian/ME-D2N_for_CDFSL.

arXiv:2210.05521공식 코드 · Papers with Code

Hybrid Inverted Index Is a Robust Accelerator for Dense Retrieval

Inverted file structure is a common technique for accelerating dense retrieval. It clusters documents based on their embeddings; during searching, it probes nearby clusters w.r.t. an input query and only evaluates documents within them by subsequent codecs, thus avoiding the expensive cost of exhaustive traversal. However, the clustering is always lossy, which results in the miss of relevant documents in the probed clusters and hence degrades retrieval quality. In contrast, lexical matching, such as overlaps of salient terms, tends to be strong feature for identifying relevant documents. In this work, we present the Hybrid Inverted Index (HI$^2$), where the embedding clusters and salient terms work collaboratively to accelerate dense retrieval. To make best of both effectiveness and efficiency, we devise a cluster selector and a term selector, to construct compact inverted lists and efficiently searching through them. Moreover, we leverage simple unsupervised algorithms as well as end-to-end knowledge distillation to learn these two modules, with the latter further boosting the effectiveness. Based on comprehensive experiments on popular retrieval benchmarks, we verify that clusters and terms indeed complement each other, enabling HI$^2$ to achieve lossless retrieval quality with competitive efficiency across various index settings. Our code and checkpoint are publicly available at https://github.com/namespace-Pt/Adon/tree/HI2.

arXiv:2210.05657

The Unreasonable Effectiveness of Fully-Connected Layers for Low-Data Regimes

Convolutional neural networks were the standard for solving many computer vision tasks until recently, when Transformers of MLP-based architectures have started to show competitive performance. These architectures typically have a vast number of weights and need to be trained on massive datasets; hence, they are not suitable for their use in low-data regimes. In this work, we propose a simple yet effective framework to improve generalization from small amounts of data. We augment modern CNNs with fully-connected (FC) layers and show the massive impact this architectural change has in low-data regimes. We further present an online joint knowledge-distillation method to utilize the extra FC layers at train time but avoid them during test time. This allows us to improve the generalization of a CNN-based model without any increase in the number of weights at test time. We perform classification experiments for a large range of network backbones and several standard datasets on supervised learning and active learning. Our experiments significantly outperform the networks without fully-connected layers, reaching a relative improvement of up to $16\%$ validation accuracy in the supervised setting without adding any extra parameters during inference.

arXiv:2210.05185공식 코드 · Papers with Code

Meta-Learning with Self-Improving Momentum Target

The idea of using a separately trained target model (or teacher) to improve the performance of the student model has been increasingly popular in various machine learning domains, and meta-learning is no exception; a recent discovery shows that utilizing task-wise target models can significantly boost the generalization performance. However, obtaining a target model for each task can be highly expensive, especially when the number of tasks for meta-learning is large. To tackle this issue, we propose a simple yet effective method, coined Self-improving Momentum Target (SiMT). SiMT generates the target model by adapting from the temporal ensemble of the meta-learner, i.e., the momentum network. This momentum network and its task-specific adaptations enjoy a favorable generalization performance, enabling self-improving of the meta-learner through knowledge distillation. Moreover, we found that perturbing parameters of the meta-learner, e.g., dropout, further stabilize this self-improving process by preventing fast convergence of the distillation loss during meta-training. Our experimental results demonstrate that SiMT brings a significant performance gain when combined with a wide range of meta-learning methods under various applications, including few-shot regression, few-shot classification, and meta-reinforcement learning. Code is available at https://github.com/jihoontack/SiMT.

arXiv:2210.04468공식 코드 · Papers with Code

Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation

Past works on multimodal machine translation (MMT) elevate bilingual setup by incorporating additional aligned vision information. However, an image-must requirement of the multimodal dataset largely hinders MMT's development -- namely that it demands an aligned form of [image, source text, target text]. This limitation is generally troublesome during the inference phase especially when the aligned image is not provided as in the normal NMT setup. Thus, in this work, we introduce IKD-MMT, a novel MMT framework to support the image-free inference phase via an inversion knowledge distillation scheme. In particular, a multimodal feature generator is executed with a knowledge distillation module, which directly generates the multimodal feature from (only) source texts as the input. While there have been a few prior works entertaining the possibility to support image-free inference for machine translation, their performances have yet to rival the image-must translation. In our experiments, we identify our method as the first image-free approach to comprehensively rival or even surpass (almost) all image-must frameworks, and achieved the state-of-the-art result on the often-used Multi30k benchmark. Our code and data are available at: https://github.com/pengr/IKD-mmt/tree/master..

arXiv:2210.04427

Asymmetric Temperature Scaling Makes Larger Networks Teach Well Again

Knowledge Distillation (KD) aims at transferring the knowledge of a well-performed neural network (the {\it teacher}) to a weaker one (the {\it student}). A peculiar phenomenon is that a more accurate model doesn't necessarily teach better, and temperature adjustment can neither alleviate the mismatched capacity. To explain this, we decompose the efficacy of KD into three parts: {\it correct guidance}, {\it smooth regularization}, and {\it class discriminability}. The last term describes the distinctness of {\it wrong class probabilities} that the teacher provides in KD. Complex teachers tend to be over-confident and traditional temperature scaling limits the efficacy of {\it class discriminability}, resulting in less discriminative wrong class probabilities. Therefore, we propose {\it Asymmetric Temperature Scaling (ATS)}, which separately applies a higher/lower temperature to the correct/wrong class. ATS enlarges the variance of wrong class probabilities in the teacher's label and makes the students grasp the absolute affinities of wrong classes to the target class as discriminative as possible. Both theoretical analysis and extensive experimental results demonstrate the effectiveness of ATS. The demo developed in Mindspore is available at \url{https://gitee.com/lxcnju/ats-mindspore} and will be available at \url{https://gitee.com/mindspore/models/tree/master/research/cv/ats}.

arXiv:2210.04208공식 코드 · Papers with Code

Let Images Give You More:Point Cloud Cross-Modal Training for Shape Analysis

Although recent point cloud analysis achieves impressive progress, the paradigm of representation learning from a single modality gradually meets its bottleneck. In this work, we take a step towards more discriminative 3D point cloud representation by fully taking advantages of images which inherently contain richer appearance information, e.g., texture, color, and shade. Specifically, this paper introduces a simple but effective point cloud cross-modality training (PointCMT) strategy, which utilizes view-images, i.e., rendered or projected 2D images of the 3D object, to boost point cloud analysis. In practice, to effectively acquire auxiliary knowledge from view images, we develop a teacher-student framework and formulate the cross modal learning as a knowledge distillation problem. PointCMT eliminates the distribution discrepancy between different modalities through novel feature and classifier enhancement criteria and avoids potential negative transfer effectively. Note that PointCMT effectively improves the point-only representation without architecture modification. Sufficient experiments verify significant gains on various datasets using appealing backbones, i.e., equipped with PointCMT, PointNet++ and PointMLP achieve state-of-the-art performance on two benchmarks, i.e., 94.4% and 86.7% accuracy on ModelNet40 and ScanObjectNN, respectively. Code will be made available at https://github.com/ZhanHeshen/PointCMT.

arXiv:2210.03885공식 코드 · Papers with Code

Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts

In this paper, we tackle the problem of domain shift. Most existing methods perform training on multiple source domains using a single model, and the same trained model is used on all unseen target domains. Such solutions are sub-optimal as each target domain exhibits its own specialty, which is not adapted. Furthermore, expecting single-model training to learn extensive knowledge from multiple source domains is counterintuitive. The model is more biased toward learning only domain-invariant features and may result in negative knowledge transfer. In this work, we propose a novel framework for unsupervised test-time adaptation, which is formulated as a knowledge distillation process to address domain shift. Specifically, we incorporate Mixture-of-Experts (MoE) as teachers, where each expert is separately trained on different source domains to maximize their specialty. Given a test-time target domain, a small set of unlabeled data is sampled to query the knowledge from MoE. As the source domains are correlated to the target domains, a transformer-based aggregator then combines the domain knowledge by examining the interconnection among them. The output is treated as a supervision signal to adapt a student prediction network toward the target domain. We further employ meta-learning to enforce the aggregator to distill positive knowledge and the student network to achieve fast adaptation. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art and validates the effectiveness of each proposed component. Our code is available at https://github.com/n3il666/Meta-DMoE.

arXiv:2210.04331

Students taught by multimodal teachers are superior action recognizers

The focal point of egocentric video understanding is modelling hand-object interactions. Standard models -- CNNs, Vision Transformers, etc. -- which receive RGB frames as input perform well, however, their performance improves further by employing additional modalities such as object detections, optical flow, audio, etc. as input. The added complexity of the required modality-specific modules, on the other hand, makes these models impractical for deployment. The goal of this work is to retain the performance of such multimodal approaches, while using only the RGB images as input at inference time. Our approach is based on multimodal knowledge distillation, featuring a multimodal teacher (in the current experiments trained only using object detections, optical flow and RGB frames) and a unimodal student (using only RGB frames as input). We present preliminary results which demonstrate that the resulting model -- distilled from a multimodal teacher -- significantly outperforms the baseline RGB model (trained without knowledge distillation), as well as an omnivorous version of itself (trained on all modalities jointly), in both standard and compositional action recognition.

arXiv:2210.04834

Knowledge Distillation Transfer Sets and their Impact on Downstream NLU Tasks

Teacher-student knowledge distillation is a popular technique for compressing today's prevailing large language models into manageable sizes that fit low-latency downstream applications. Both the teacher and the choice of transfer set used for distillation are crucial ingredients in creating a high quality student. Yet, the generic corpora used to pretrain the teacher and the corpora associated with the downstream target domain are often significantly different, which raises a natural question: should the student be distilled over the generic corpora, so as to learn from high-quality teacher predictions, or over the downstream task corpora to align with finetuning? Our study investigates this trade-off using Domain Classification (DC) and Intent Classification/Named Entity Recognition (ICNER) as downstream tasks. We distill several multilingual students from a larger multilingual LM with varying proportions of generic and task-specific datasets, and report their performance after finetuning on DC and ICNER. We observe significant improvements across tasks and test sets when only task-specific corpora is used. We also report on how the impact of adding task-specific data to the transfer set correlates with the similarity between generic and task-specific data. Our results clearly indicate that, while distillation from a generic LM benefits downstream tasks, students learn better using target domain data even if it comes at the price of noisier teacher predictions. In other words, target domain data still trumps teacher knowledge.

arXiv:2210.03488공식 코드 · Papers with Code

AlphaFold Distillation for Protein Design

Inverse protein folding, the process of designing sequences that fold into a specific 3D structure, is crucial in bio-engineering and drug discovery. Traditional methods rely on experimentally resolved structures, but these cover only a small fraction of protein sequences. Forward folding models like AlphaFold offer a potential solution by accurately predicting structures from sequences. However, these models are too slow for integration into the optimization loop of inverse folding models during training. To address this, we propose using knowledge distillation on folding model confidence metrics, such as pTM or pLDDT scores, to create a faster and end-to-end differentiable distilled model. This model can then be used as a structure consistency regularizer in training the inverse folding model. Our technique is versatile and can be applied to other design tasks, such as sequence-based protein infilling. Experimental results show that our method outperforms non-regularized baselines, yielding up to 3% improvement in sequence recovery and up to 45% improvement in protein diversity while maintaining structural consistency in generated sequences. Code is available at https://github.com/IBM/AFDistill

arXiv:2210.03459

Mutual Learning of Single- and Multi-Channel End-to-End Neural Diarization

Due to the high performance of multi-channel speech processing, we can use the outputs from a multi-channel model as teacher labels when training a single-channel model with knowledge distillation. To the contrary, it is also known that single-channel speech data can benefit multi-channel models by mixing it with multi-channel speech data during training or by using it for model pretraining. This paper focuses on speaker diarization and proposes to conduct the above bi-directional knowledge transfer alternately. We first introduce an end-to-end neural diarization model that can handle both single- and multi-channel inputs. Using this model, we alternately conduct i) knowledge distillation from a multi-channel model to a single-channel model and ii) finetuning from the distilled single-channel model to a multi-channel model. Experimental results on two-speaker data show that the proposed method mutually improved single- and multi-channel speaker diarization performances.

arXiv:2210.03477공식 코드 · Papers with Code

IDa-Det: An Information Discrepancy-aware Distillation for 1-bit Detectors

Knowledge distillation (KD) has been proven to be useful for training compact object detection models. However, we observe that KD is often effective when the teacher model and student counterpart share similar proposal information. This explains why existing KD methods are less effective for 1-bit detectors, caused by a significant information discrepancy between the real-valued teacher and the 1-bit student. This paper presents an Information Discrepancy-aware strategy (IDa-Det) to distill 1-bit detectors that can effectively eliminate information discrepancies and significantly reduce the performance gap between a 1-bit detector and its real-valued counterpart. We formulate the distillation process as a bi-level optimization formulation. At the inner level, we select the representative proposals with maximum information discrepancy. We then introduce a novel entropy distillation loss to reduce the disparity based on the selected proposals. Extensive experiments demonstrate IDa-Det's superiority over state-of-the-art 1-bit detectors and KD methods on both PASCAL VOC and COCO datasets. IDa-Det achieves a 76.9% mAP for a 1-bit Faster-RCNN with ResNet-18 backbone. Our code is open-sourced on https://github.com/SteveTsui/IDa-Det.

arXiv:2210.03664공식 코드 · Papers with Code

Bi-directional Weakly Supervised Knowledge Distillation for Whole Slide Image Classification

Computer-aided pathology diagnosis based on the classification of Whole Slide Image (WSI) plays an important role in clinical practice, and it is often formulated as a weakly-supervised Multiple Instance Learning (MIL) problem. Existing methods solve this problem from either a bag classification or an instance classification perspective. In this paper, we propose an end-to-end weakly supervised knowledge distillation framework (WENO) for WSI classification, which integrates a bag classifier and an instance classifier in a knowledge distillation framework to mutually improve the performance of both classifiers. Specifically, an attention-based bag classifier is used as the teacher network, which is trained with weak bag labels, and an instance classifier is used as the student network, which is trained using the normalized attention scores obtained from the teacher network as soft pseudo labels for the instances in positive bags. An instance feature extractor is shared between the teacher and the student to further enhance the knowledge exchange between them. In addition, we propose a hard positive instance mining strategy based on the output of the student network to force the teacher network to keep mining hard positive instances. WENO is a plug-and-play framework that can be easily applied to any existing attention-based bag classification methods. Extensive experiments on five datasets demonstrate the efficiency of WENO. Code is available at https://github.com/miccaiif/WENO.

arXiv:2210.03625공식 코드 · Papers with Code

C2KD: Cross-Lingual Cross-Modal Knowledge Distillation for Multilingual Text-Video Retrieval

Multilingual text-video retrieval methods have improved significantly in recent years, but the performance for other languages lags behind English. We propose a Cross-Lingual Cross-Modal Knowledge Distillation method to improve multilingual text-video retrieval. Inspired by the fact that English text-video retrieval outperforms other languages, we train a student model using input text in different languages to match the cross-modal predictions from teacher models using input text in English. We propose a cross entropy based objective which forces the distribution over the student's text-video similarity scores to be similar to those of the teacher models. We introduce a new multilingual video dataset, Multi-YouCook2, by translating the English captions in the YouCook2 video dataset to 8 other languages. Our method improves multilingual text-video retrieval performance on Multi-YouCook2 and several other datasets such as Multi-MSRVTT and VATEX. We also conducted an analysis on the effectiveness of different multilingual text models as teachers. The code, models, and dataset are available at https://github.com/roudimit/c2kd.

arXiv:2210.02808공식 코드 · Papers with Code

Effective Self-supervised Pre-training on Low-compute Networks without Distillation

Despite the impressive progress of self-supervised learning (SSL), its applicability to low-compute networks has received limited attention. Reported performance has trailed behind standard supervised pre-training by a large margin, barring self-supervised learning from making an impact on models that are deployed on device. Most prior works attribute this poor performance to the capacity bottleneck of the low-compute networks and opt to bypass the problem through the use of knowledge distillation (KD). In this work, we revisit SSL for efficient neural networks, taking a closer at what are the detrimental factors causing the practical limitations, and whether they are intrinsic to the self-supervised low-compute setting. We find that, contrary to accepted knowledge, there is no intrinsic architectural bottleneck, we diagnose that the performance bottleneck is related to the model complexity vs regularization strength trade-off. In particular, we start by empirically observing that the use of local views can have a dramatic impact on the effectiveness of the SSL methods. This hints at view sampling being one of the performance bottlenecks for SSL on low-capacity networks. We hypothesize that the view sampling strategy for large neural networks, which requires matching views in very diverse spatial scales and contexts, is too demanding for low-capacity architectures. We systematize the design of the view sampling mechanism, leading to a new training methodology that consistently improves the performance across different SSL methods (e.g. MoCo-v2, SwAV, DINO), different low-size networks (e.g. MobileNetV2, ResNet18, ResNet34, ViT-Ti), and different tasks (linear probe, object detection, instance segmentation and semi-supervised learning). Our best models establish a new state-of-the-art for SSL methods on low-compute networks despite not using a KD loss term.

arXiv:2210.03114공식 코드 · Papers with Code

CLIP model is an Efficient Continual Learner

The continual learning setting aims to learn new tasks over time without forgetting the previous ones. The literature reports several significant efforts to tackle this problem with limited or no access to previous task data. Among such efforts, typical solutions offer sophisticated techniques involving memory replay, knowledge distillation, model regularization, and dynamic network expansion. The resulting methods have a retraining cost at each learning task, dedicated memory requirements, and setting-specific design choices. In this work, we show that a frozen CLIP (Contrastive Language-Image Pretraining) model offers astounding continual learning performance without any fine-tuning (zero-shot evaluation). We evaluate CLIP under a variety of settings including class-incremental, domain-incremental and task-agnostic incremental learning on five popular benchmarks (ImageNet-100 & 1K, CORe50, CIFAR-100, and TinyImageNet). Without any bells and whistles, the CLIP model outperforms the state-of-the-art continual learning approaches in the majority of the settings. We show the effect on the CLIP model's performance by varying text inputs with simple prompt templates. To the best of our knowledge, this is the first work to report the CLIP zero-shot performance in a continual setting. We advocate the use of this strong yet embarrassingly simple baseline for future comparisons in the continual learning tasks.

arXiv:2210.02099

Automated Graph Self-supervised Learning via Multi-teacher Knowledge Distillation

Self-supervised learning on graphs has recently achieved remarkable success in graph representation learning. With hundreds of self-supervised pretext tasks proposed over the past few years, the research community has greatly developed, and the key is no longer to design more powerful but complex pretext tasks, but to make more effective use of those already on hand. This paper studies the problem of how to automatically, adaptively, and dynamically learn instance-level self-supervised learning strategies for each node from a given pool of pretext tasks. In this paper, we propose a novel multi-teacher knowledge distillation framework for Automated Graph Self-Supervised Learning (AGSSL), which consists of two main branches: (i) Knowledge Extraction: training multiple teachers with different pretext tasks, so as to extract different levels of knowledge with different inductive biases; (ii) Knowledge Integration: integrating different levels of knowledge and distilling them into the student model. Without simply treating different teachers as equally important, we provide a provable theoretical guideline for how to integrate the knowledge of different teachers, i.e., the integrated teacher probability should be close to the true Bayesian class-probability. To approach the theoretical optimum in practice, two adaptive knowledge integration strategies are proposed to construct a relatively "good" integrated teacher. Extensive experiments on eight datasets show that AGSSL can benefit from multiple pretext tasks, outperforming the corresponding individual tasks; by combining a few simple but classical pretext tasks, the resulting performance is comparable to other leading counterparts.

arXiv:2210.01973

Meta-Ensemble Parameter Learning

Ensemble of machine learning models yields improved performance as well as robustness. However, their memory requirements and inference costs can be prohibitively high. Knowledge distillation is an approach that allows a single model to efficiently capture the approximate performance of an ensemble while showing poor scalability as demand for re-training when introducing new teacher models. In this paper, we study if we can utilize the meta-learning strategy to directly predict the parameters of a single model with comparable performance of an ensemble. Hereto, we introduce WeightFormer, a Transformer-based model that can predict student network weights layer by layer in a forward pass, according to the teacher model parameters. The proprieties of WeightFormer are investigated on the CIFAR-10, CIFAR-100, and ImageNet datasets for model structures of VGGNet-11, ResNet-50, and ViT-B/32, where it demonstrates that our method can achieve approximate classification performance of an ensemble and outperforms both the single network and standard knowledge distillation. More encouragingly, we show that WeightFormer results can further exceeds average ensemble with minor fine-tuning. Importantly, our task along with the model and results can potentially lead to a new, more efficient, and scalable paradigm of ensemble networks parameter learning.

arXiv:2210.02190

Domain Discrepancy Aware Distillation for Model Aggregation in Federated Learning

Knowledge distillation has recently become popular as a method of model aggregation on the server for federated learning. It is generally assumed that there are abundant public unlabeled data on the server. However, in reality, there exists a domain discrepancy between the datasets of the server domain and a client domain, which limits the performance of knowledge distillation. How to improve the aggregation under such a domain discrepancy setting is still an open problem. In this paper, we first analyze the generalization bound of the aggregation model produced from knowledge distillation for the client domains, and then describe two challenges, server-to-client discrepancy and client-to-client discrepancy, brought to the aggregation model by the domain discrepancies. Following our analysis, we propose an adaptive knowledge aggregation algorithm FedD3A based on domain discrepancy aware distillation to lower the bound. FedD3A performs adaptive weighting at the sample level in each round of FL. For each sample in the server domain, only the client models of its similar domains will be selected for playing the teacher role. To achieve this, we show that the discrepancy between the server-side sample and the client domain can be approximately measured using a subspace projection matrix calculated on each client without accessing its raw data. The server can thus leverage the projection matrices from multiple clients to assign weights to the corresponding teacher models for each server-side sample. We validate FedD3A on two popular cross-domain datasets and show that it outperforms the compared competitors in both cross-silo and cross-device FL settings.