PyTorchKR Papers

최신 논문

총 10,163편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2107.11979

HYPER-SNN: Towards Energy-efficient Quantized Deep Spiking Neural Networks for Hyperspectral Image Classification

Hyper spectral images (HSI) provide rich spectral and spatial information across a series of contiguous spectral bands. However, the accurate processing of the spectral and spatial correlation between the bands requires the use of energy-expensive 3-D Convolutional Neural Networks (CNNs). To address this challenge, we propose the use of Spiking Neural Networks (SNNs) that are generated from iso-architecture CNNs and trained with quantization-aware gradient descent to optimize their weights, membrane leak, and firing thresholds. During both training and inference, the analog pixel values of a HSI are directly applied to the input layer of the SNN without the need to convert to a spike-train. The reduced latency of our training technique combined with high activation sparsity yields significant improvements in computational efficiency. We evaluate our proposal using three HSI datasets on a 3-D and a 3-D/2-D hybrid convolutional architecture. We achieve overall accuracy, average accuracy, and kappa coefficient of 98.68%, 98.34%, and 98.20% respectively with 5 time steps (inference latency) and 6-bit weight quantization on the Indian Pines dataset. In particular, our models achieved accuracies similar to state-of-the-art (SOTA) with 560.6 and 44.8 times less compute energy on average over three HSI datasets than an iso-architecture full-precision and 6-bit quantized CNN, respectively.

arXiv:2107.11921

Compensation Learning

Weighting strategy prevails in machine learning. For example, a common approach in robust machine learning is to exert lower weights on samples which are likely to be noisy or quite hard. This study reveals another undiscovered strategy, namely, compensating. Various incarnations of compensating have been utilized but it has not been explicitly revealed. Learning with compensating is called compensation learning and a systematic taxonomy is constructed for it in this study. In our taxonomy, compensation learning is divided on the basis of the compensation targets, directions, inference manners, and granularity levels. Many existing learning algorithms including some classical ones can be viewed or understood at least partially as compensation techniques. Furthermore, a family of new learning algorithms can be obtained by plugging the compensation learning into existing learning algorithms. Specifically, two concrete new learning algorithms are proposed for robust machine learning. Extensive experiments on image classification and text sentiment analysis verify the effectiveness of the two new algorithms. Compensation learning can also be used in other various learning scenarios, such as imbalance learning, clustering, regression, and so on.

arXiv:2107.11787공식 코드 · Papers with Code

Leveraging Auxiliary Tasks with Affinity Learning for Weakly Supervised Semantic Segmentation

Semantic segmentation is a challenging task in the absence of densely labelled data. Only relying on class activation maps (CAM) with image-level labels provides deficient segmentation supervision. Prior works thus consider pre-trained models to produce coarse saliency maps to guide the generation of pseudo segmentation labels. However, the commonly used off-line heuristic generation process cannot fully exploit the benefits of these coarse saliency maps. Motivated by the significant inter-task correlation, we propose a novel weakly supervised multi-task framework termed as AuxSegNet, to leverage saliency detection and multi-label image classification as auxiliary tasks to improve the primary task of semantic segmentation using only image-level ground-truth labels. Inspired by their similar structured semantics, we also propose to learn a cross-task global pixel-level affinity map from the saliency and segmentation representations. The learned cross-task affinity can be used to refine saliency predictions and propagate CAM maps to provide improved pseudo labels for both tasks. The mutual boost between pseudo label updating and cross-task affinity learning enables iterative improvements on segmentation performance. Extensive experiments demonstrate the effectiveness of the proposed auxiliary learning network structure and the cross-task affinity learning method. The proposed approach achieves state-of-the-art weakly supervised segmentation performance on the challenging PASCAL VOC 2012 and MS COCO benchmarks.

arXiv:2107.11711

Temporal-wise Attention Spiking Neural Networks for Event Streams Classification

How to effectively and efficiently deal with spatio-temporal event streams, where the events are generally sparse and non-uniform and have the microsecond temporal resolution, is of great value and has various real-life applications. Spiking neural network (SNN), as one of the brain-inspired event-triggered computing models, has the potential to extract effective spatio-temporal features from the event streams. However, when aggregating individual events into frames with a new higher temporal resolution, existing SNN models do not attach importance to that the serial frames have different signal-to-noise ratios since event streams are sparse and non-uniform. This situation interferes with the performance of existing SNNs. In this work, we propose a temporal-wise attention SNN (TA-SNN) model to learn frame-based representation for processing event streams. Concretely, we extend the attention concept to temporal-wise input to judge the significance of frames for the final decision at the training stage, and discard the irrelevant frames at the inference stage. We demonstrate that TA-SNN models improve the accuracy of event streams classification tasks. We also study the impact of multiple-scale temporal resolutions for frame-based representation. Our approach is tested on three different classification tasks: gesture recognition, image classification, and spoken digit recognition. We report the state-of-the-art results on these tasks, and get the essential improvement of accuracy (almost 19\%) for gesture recognition with only 60 ms.

arXiv:2107.11707

Boosting Video Captioning with Dynamic Loss Network

Video captioning is one of the challenging problems at the intersection of vision and language, having many real-life applications in video retrieval, video surveillance, assisting visually challenged people, Human-machine interface, and many more. Recent deep learning based methods have shown promising results but are still on the lower side than other vision tasks (such as image classification, object detection). A significant drawback with existing video captioning methods is that they are optimized over cross-entropy loss function, which is uncorrelated to the de facto evaluation metrics (BLEU, METEOR, CIDER, ROUGE). In other words, cross-entropy is not a proper surrogate of the true loss function for video captioning. To mitigate this, methods like REINFORCE, Actor-Critic, and Minimum Risk Training (MRT) have been applied but have limitations and are not very effective. This paper proposes an alternate solution by introducing a dynamic loss network (DLN), providing an additional feedback signal that reflects the evaluation metrics directly. Our solution proves to be more efficient than other solutions and can be easily adapted to similar tasks. Our results on Microsoft Research Video Description Corpus (MSVD) and MSR-Video to Text (MSRVTT) datasets outperform previous methods.

arXiv:2107.11643

An Uncertainty-Aware Deep Learning Framework for Defect Detection in Casting Products

Defects are unavoidable in casting production owing to the complexity of the casting process. While conventional human-visual inspection of casting products is slow and unproductive in mass productions, an automatic and reliable defect detection not just enhances the quality control process but positively improves productivity. However, casting defect detection is a challenging task due to diversity and variation in defects' appearance. Convolutional neural networks (CNNs) have been widely applied in both image classification and defect detection tasks. Howbeit, CNNs with frequentist inference require a massive amount of data to train on and still fall short in reporting beneficial estimates of their predictive uncertainty. Accordingly, leveraging the transfer learning paradigm, we first apply four powerful CNN-based models (VGG16, ResNet50, DenseNet121, and InceptionResNetV2) on a small dataset to extract meaningful features. Extracted features are then processed by various machine learning algorithms to perform the classification task. Simulation results demonstrate that linear support vector machine (SVM) and multi-layer perceptron (MLP) show the finest performance in defect detection of casting images. Secondly, to achieve a reliable classification and to measure epistemic uncertainty, we employ an uncertainty quantification (UQ) technique (ensemble of MLP models) using features extracted from four pre-trained CNNs. UQ confusion matrix and uncertainty accuracy metric are also utilized to evaluate the predictive uncertainty estimates. Comprehensive comparisons reveal that UQ method based on VGG16 outperforms others to fetch uncertainty. We believe an uncertainty-aware automatic defect detection solution will reinforce casting productions quality assurance.

arXiv:2107.11626

Multi-Label Image Classification with Contrastive Learning

Recently, as an effective way of learning latent representations, contrastive learning has been increasingly popular and successful in various domains. The success of constrastive learning in single-label classifications motivates us to leverage this learning framework to enhance distinctiveness for better performance in multi-label image classification. In this paper, we show that a direct application of contrastive learning can hardly improve in multi-label cases. Accordingly, we propose a novel framework for multi-label classification with contrastive learning in a fully supervised setting, which learns multiple representations of an image under the context of different labels. This facilities a simple yet intuitive adaption of contrastive learning into our model to boost its performance in multi-label image classification. Extensive experiments on two benchmark datasets show that the proposed framework achieves state-of-the-art performance in the comparison with the advanced methods in multi-label classification.

arXiv:2107.11574

Reconstructing Images of Two Adjacent Objects through Scattering Medium Using Generative Adversarial Network

Reconstruction of image by using convolutional neural networks (CNNs) has been vigorously studied in the last decade. Until now, there have being developed several techniques for imaging of a single object through scattering medium by using neural networks, however how to reconstruct images of more than one object simultaneously seems hard to realize. In this paper, we demonstrate an approach by using generative adversarial network (GAN) to reconstruct images of two adjacent objects through scattering media. We construct an imaging system for imaging of two adjacent objects behind the scattering media. In general, as the light field of two adjacent object images pass through the scattering slab, a speckle pattern is obtained. The designed adversarial network, which is called as YGAN, is employed to reconstruct the images simultaneously. It is shown that based on the trained YGAN, we can reconstruct images of two adjacent objects from one speckle pattern with high fidelity. In addition, we study the influence of the object image types, and the distance between the two adjacent objects on the fidelity of the reconstructed images. Moreover even if another scattering medium is inserted between the two objects, we can also reconstruct the images of two objects from a speckle with high quality. The technique presented in this work can be used for applications in areas of medical image analysis, such as medical image classification, segmentation, and studies of multi-object scattering imaging etc.

arXiv:2107.12292공식 코드 · Papers with Code

Contextual Transformer Networks for Visual Recognition

Transformer with self-attention has led to the revolutionizing of natural language processing field, and recently inspires the emergence of Transformer-style architecture design with competitive results in numerous computer vision tasks. Nevertheless, most of existing designs directly employ self-attention over a 2D feature map to obtain the attention matrix based on pairs of isolated queries and keys at each spatial location, but leave the rich contexts among neighbor keys under-exploited. In this work, we design a novel Transformer-style module, i.e., Contextual Transformer (CoT) block, for visual recognition. Such design fully capitalizes on the contextual information among input keys to guide the learning of dynamic attention matrix and thus strengthens the capacity of visual representation. Technically, CoT block first contextually encodes input keys via a $3\times3$ convolution, leading to a static contextual representation of inputs. We further concatenate the encoded keys with input queries to learn the dynamic multi-head attention matrix through two consecutive $1\times1$ convolutions. The learnt attention matrix is multiplied by input values to achieve the dynamic contextual representation of inputs. The fusion of the static and dynamic contextual representations are finally taken as outputs. Our CoT block is appealing in the view that it can readily replace each $3\times3$ convolution in ResNet architectures, yielding a Transformer-style backbone named as Contextual Transformer Networks (CoTNet). Through extensive experiments over a wide range of applications (e.g., image recognition, object detection and instance segmentation), we validate the superiority of CoTNet as a stronger backbone. Source code is available at \url{https://github.com/JDAI-CV/CoTNet}.

arXiv:2107.12028공식 코드 · Papers with Code

Parametric Contrastive Learning

In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnable centers to rebalance from an optimization perspective. Further, we analyze our PaCo loss under a balanced setting. Our analysis demonstrates that PaCo can adaptively enhance the intensity of pushing samples of the same class close as more samples are pulled together with their corresponding centers and benefit hard example learning. Experiments on long-tailed CIFAR, ImageNet, Places, and iNaturalist 2018 manifest the new state-of-the-art for long-tailed recognition. On full ImageNet, models trained with PaCo loss surpass supervised contrastive learning across various ResNet backbones, e.g., our ResNet-200 achieves 81.8% top-1 accuracy. Our code is available at https://github.com/dvlab-research/Parametric-Contrastive-Learning.

arXiv:2107.11817

Go Wider Instead of Deeper

More transformer blocks with residual connections have recently achieved impressive results on various tasks. To achieve better performance with fewer trainable parameters, recent methods are proposed to go shallower by parameter sharing or model compressing along with the depth. However, weak modeling capacity limits their performance. Contrastively, going wider by inducing more trainable matrixes and parameters would produce a huge model requiring advanced parallelism to train and inference. In this paper, we propose a parameter-efficient framework, going wider instead of deeper. Specially, following existing works, we adapt parameter sharing to compress along depth. But, such deployment would limit the performance. To maximize modeling capacity, we scale along model width by replacing feed-forward network (FFN) with mixture-of-experts (MoE). Across transformer blocks, instead of sharing normalization layers, we propose to use individual layernorms to transform various semantic representations in a more parameter-efficient way. To evaluate our plug-and-run framework, we design WideNet and conduct comprehensive experiments on popular computer vision and natural language processing benchmarks. On ImageNet-1K, our best model outperforms Vision Transformer (ViT) by $1.5\%$ with $0.72 \times$ trainable parameters. Using $0.46 \times$ and $0.13 \times$ parameters, our WideNet can still surpass ViT and ViT-MoE by $0.8\%$ and $2.1\%$, respectively. On four natural language processing datasets, WideNet outperforms ALBERT by $1.8\%$ on average and surpass BERT using factorized embedding parameterization by $0.8\%$ with fewer parameters.

arXiv:2107.12329공식 코드 · Papers with Code

AASAE: Augmentation-Augmented Stochastic Autoencoders

Recent methods for self-supervised learning can be grouped into two paradigms: contrastive and non-contrastive approaches. Their success can largely be attributed to data augmentation pipelines which generate multiple views of a single input that preserve the underlying semantics. In this work, we introduce augmentation-augmented stochastic autoencoders (AASAE), yet another alternative to self-supervised learning, based on autoencoding. We derive AASAE starting from the conventional variational autoencoder (VAE), by replacing the KL divergence regularization, which is agnostic to the input domain, with data augmentations that explicitly encourage the internal representations to encode domain-specific invariances and equivariances. We empirically evaluate the proposed AASAE on image classification, similar to how recent contrastive and non-contrastive learning algorithms have been evaluated. Our experiments confirm the effectiveness of data augmentation as a replacement for KL divergence regularization. The AASAE outperforms the VAE by 30% on CIFAR-10, 40% on STL-10 and 45% on Imagenet. On CIFAR-10 and STL-10, the results for AASAE are largely comparable to the state-of-the-art algorithms for self-supervised learning.

arXiv:2107.11001

Photon-Starved Scene Inference using Single Photon Cameras

Scene understanding under low-light conditions is a challenging problem. This is due to the small number of photons captured by the camera and the resulting low signal-to-noise ratio (SNR). Single-photon cameras (SPCs) are an emerging sensing modality that are capable of capturing images with high sensitivity. Despite having minimal read-noise, images captured by SPCs in photon-starved conditions still suffer from strong shot noise, preventing reliable scene inference. We propose photon scale-space a collection of high-SNR images spanning a wide range of photons-per-pixel (PPP) levels (but same scene content) as guides to train inference model on low photon flux images. We develop training techniques that push images with different illumination levels closer to each other in feature representation space. The key idea is that having a spectrum of different brightness levels during training enables effective guidance, and increases robustness to shot noise even in extreme noise cases. Based on the proposed approach, we demonstrate, via simulations and real experiments with a SPAD camera, high-performance on various inference tasks such as image classification and monocular depth estimation under ultra low-light, down to < 1 PPP.

arXiv:2107.10998

Pruning Ternary Quantization

Inference time, model size, and accuracy are three key factors in deep model compression. Most of the existing work addresses these three key factors separately as it is difficult to optimize them all at the same time. For example, low-bit quantization aims at obtaining a faster model; weight sharing quantization aims at improving compression ratio and accuracy; and mixed-precision quantization aims at balancing accuracy and inference time. To simultaneously optimize bit-width, model size, and accuracy, we propose pruning ternary quantization (PTQ): a simple, effective, symmetric ternary quantization method. We integrate L2 normalization, pruning, and the weight decay term to reduce the weight discrepancy in the gradient estimator during quantization, thus producing highly compressed ternary weights. Our method brings the highest test accuracy and the highest compression ratio. For example, it produces a 939kb (49$\times$) 2bit ternary ResNet-18 model with only 4\% accuracy drop on the ImageNet dataset. It compresses 170MB Mask R-CNN to 5MB (34$\times$) with only 2.8\% average precision drop. Our method is verified on image classification, object detection/segmentation tasks with different network structures such as ResNet-18, ResNet-50, and MobileNetV2.

arXiv:2107.11170공식 코드 · Papers with Code

Bias Loss for Mobile Neural Networks

Compact convolutional neural networks (CNNs) have witnessed exceptional improvements in performance in recent years. However, they still fail to provide the same predictive power as CNNs with a large number of parameters. The diverse and even abundant features captured by the layers is an important characteristic of these successful CNNs. However, differences in this characteristic between large CNNs and their compact counterparts have rarely been investigated. In compact CNNs, due to the limited number of parameters, abundant features are unlikely to be obtained, and feature diversity becomes an essential characteristic. Diverse features present in the activation maps derived from a data point during model inference may indicate the presence of a set of unique descriptors necessary to distinguish between objects of different classes. In contrast, data points with low feature diversity may not provide a sufficient amount of unique descriptors to make a valid prediction; we refer to them as random predictions. Random predictions can negatively impact the optimization process and harm the final performance. This paper proposes addressing the problem raised by random predictions by reshaping the standard cross-entropy to make it biased toward data points with a limited number of unique descriptive features. Our novel Bias Loss focuses the training on a set of valuable data points and prevents the vast number of samples with poor learning features from misleading the optimization process. Furthermore, to show the importance of diversity, we present a family of SkipNet models whose architectures are brought to boost the number of unique descriptors in the last layers. Our Skipnet-M can achieve 1% higher classification accuracy than MobileNetV3 Large.

arXiv:2107.11099

RGB Image Classification with Quantum Convolutional Ansaetze

With the rapid growth of qubit numbers and coherence times in quantum hardware technology, implementing shallow neural networks on the so-called Noisy Intermediate-Scale Quantum (NISQ) devices has attracted a lot of interest. Many quantum (convolutional) circuit ansaetze are proposed for grayscale images classification tasks with promising empirical results. However, when applying these ansaetze on RGB images, the intra-channel information that is useful for vision tasks is not extracted effectively. In this paper, we propose two types of quantum circuit ansaetze to simulate convolution operations on RGB images, which differ in the way how inter-channel and intra-channel information are extracted. To the best of our knowledge, this is the first work of a quantum convolutional circuit to deal with RGB images effectively, with a higher test accuracy compared to the purely classical CNNs. We also investigate the relationship between the size of quantum circuit ansatz and the learnability of the hybrid quantum-classical convolutional neural network. Through experiments based on CIFAR-10 and MNIST datasets, we demonstrate that a larger size of the quantum circuit ansatz improves predictive performance in multiclass classification tasks, providing useful insights for near term quantum algorithm developments.

arXiv:2107.11049공식 코드 · Papers with Code

MCDAL: Maximum Classifier Discrepancy for Active Learning

Recent state-of-the-art active learning methods have mostly leveraged Generative Adversarial Networks (GAN) for sample acquisition; however, GAN is usually known to suffer from instability and sensitivity to hyper-parameters. In contrast to these methods, we propose in this paper a novel active learning framework that we call Maximum Classifier Discrepancy for Active Learning (MCDAL) which takes the prediction discrepancies between multiple classifiers. In particular, we utilize two auxiliary classification layers that learn tighter decision boundaries by maximizing the discrepancies among them. Intuitively, the discrepancies in the auxiliary classification layers' predictions indicate the uncertainty in the prediction. In this regard, we propose a novel method to leverage the classifier discrepancies for the acquisition function for active learning. We also provide an interpretation of our idea in relation to existing GAN based active learning methods and domain adaptation frameworks. Moreover, we empirically demonstrate the utility of our approach where the performance of our approach exceeds the state-of-the-art methods on several image classification and semantic segmentation datasets in active learning setups.

arXiv:2107.11359

Rethinking Hard-Parameter Sharing in Multi-Domain Learning

Hard parameter sharing in multi-domain learning (MDL) allows domains to share some of the model parameters to reduce storage cost while improving prediction accuracy. One common sharing practice is to share the bottom layers of a deep neural network among domains while using separate top layers for each domain. In this work, we revisit this common practice via an empirical study on image classification tasks from a diverse set of visual domains and make two surprising observations. (1) Using separate bottom-layer parameters could achieve significantly better performance than the common practice and this phenomenon holds with different experimental settings. (2) A multi-domain model with a small proportion of domain-specific parameters from bottom layers can achieve competitive performance with independent models trained on each domain separately. Our observations suggest that people adopt the new strategy of using separate bottom-layer parameters as a stronger baseline for model design in MDL.

arXiv:2107.10976

Federated Learning Versus Classical Machine Learning: A Convergence Comparison

In the past few decades, machine learning has revolutionized data processing for large scale applications. Simultaneously, increasing privacy threats in trending applications led to the redesign of classical data training models. In particular, classical machine learning involves centralized data training, where the data is gathered, and the entire training process executes at the central server. Despite significant convergence, this training involves several privacy threats on participants' data when shared with the central cloud server. To this end, federated learning has achieved significant importance over distributed data training. In particular, the federated learning allows participants to collaboratively train the local models on local data without revealing their sensitive information to the central cloud server. In this paper, we perform a convergence comparison between classical machine learning and federated learning on two publicly available datasets, namely, logistic-regression-MNIST dataset and image-classification-CIFAR-10 dataset. The simulation results demonstrate that federated learning achieves higher convergence within limited communication rounds while maintaining participants' anonymity. We hope that this research will show the benefits and help federated learning to be implemented widely.

arXiv:2107.10963

Compositional Models: Multi-Task Learning and Knowledge Transfer with Modular Networks

Conditional computation and modular networks have been recently proposed for multitask learning and other problems as a way to decompose problem solving into multiple reusable computational blocks. We propose a new approach for learning modular networks based on the isometric version of ResNet with all residual blocks having the same configuration and the same number of parameters. This architectural choice allows adding, removing and changing the order of residual blocks. In our method, the modules can be invoked repeatedly and allow knowledge transfer to novel tasks by adjusting the order of computation. This allows soft weight sharing between tasks with only a small increase in the number of parameters. We show that our method leads to interpretable self-organization of modules in case of multi-task learning, transfer learning and domain adaptation while achieving competitive results on those tasks. From practical perspective, our approach allows to: (a) reuse existing modules for learning new task by adjusting the computation order, (b) use it for unsupervised multi-source domain adaptation to illustrate that adaptation to unseen data can be achieved by only manipulating the order of pretrained modules, (c) show how our approach can be used to increase accuracy of existing architectures for image classification tasks such as ImageNet, without any parameter increase, by reusing the same block multiple times.