PyTorchKR Papers

최신 논문

총 10,163편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2010.12747

Scale-, shift- and rotation-invariant diffractive optical networks

Recent research efforts in optical computing have gravitated towards developing optical neural networks that aim to benefit from the processing speed and parallelism of optics/photonics in machine learning applications. Among these endeavors, Diffractive Deep Neural Networks (D2NNs) harness light-matter interaction over a series of trainable surfaces, designed using deep learning, to compute a desired statistical inference task as the light waves propagate from the input plane to the output field-of-view. Although, earlier studies have demonstrated the generalization capability of diffractive optical networks to unseen data, achieving e.g., >98% image classification accuracy for handwritten digits, these previous designs are in general sensitive to the spatial scaling, translation and rotation of the input objects. Here, we demonstrate a new training strategy for diffractive networks that introduces input object translation, rotation and/or scaling during the training phase as uniformly distributed random variables to build resilience in their blind inference performance against such object transformations. This training strategy successfully guides the evolution of the diffractive optical network design towards a solution that is scale-, shift- and rotation-invariant, which is especially important and useful for dynamic machine vision applications in e.g., autonomous cars, in-vivo imaging of biomedical specimen, among others.

arXiv:2010.13116

An empirical study of domain-agnostic semi-supervised learning via energy-based models: joint-training and pre-training

A class of recent semi-supervised learning (SSL) methods heavily rely on domain-specific data augmentations. In contrast, generative SSL methods involve unsupervised learning based on generative models by either joint-training or pre-training, and are more appealing from the perspective of being domain-agnostic, since they do not inherently require data augmentations. Joint-training estimates the joint distribution of observations and labels, while pre-training is taken over observations only. Recently, energy-based models (EBMs) have achieved promising results for generative modeling. Joint-training via EBMs for SSL has been explored with encouraging results across different data modalities. In this paper, we make two contributions. First, we explore pre-training via EBMs for SSL and compare it to joint-training. Second, a suite of experiments are conducted over domains of image classification and natural language labeling to give a realistic whole picture of the performances of EBM based SSL methods. It is found that joint-training EBMs outperform pre-training EBMs marginally but nearly consistently.

arXiv:2010.12993

Multi-task Supervised Learning via Cross-learning

In this paper we consider a problem known as multi-task learning, consisting of fitting a set of classifier or regression functions intended for solving different tasks. In our novel formulation, we couple the parameters of these functions, so that they learn in their task specific domains while staying close to each other. This facilitates cross-fertilization in which data collected across different domains help improving the learning performance at each other task. First, we present a simplified case in which the goal is to estimate the means of two Gaussian variables, for the purpose of gaining some insights on the advantage of the proposed cross-learning strategy. Then we provide a stochastic projected gradient algorithm to perform cross-learning over a generic loss function. If the number of parameters is large, then the projection step becomes computationally expensive. To avoid this situation, we derive a primal-dual algorithm that exploits the structure of the dual problem, achieving a formulation whose complexity only depends on the number of tasks. Preliminary numerical experiments for image classification by neural networks trained on a dataset divided in different domains corroborate that the cross-learned function outperforms both the task-specific and the consensus approaches.

arXiv:2010.13498공식 코드 · Papers with Code

Scalable Bayesian neural networks by layer-wise input augmentation

We introduce implicit Bayesian neural networks, a simple and scalable approach for uncertainty representation in deep learning. Standard Bayesian approach to deep learning requires the impractical inference of the posterior distribution over millions of parameters. Instead, we propose to induce a distribution that captures the uncertainty over neural networks by augmenting each layer's inputs with latent variables. We present appropriate input distributions and demonstrate state-of-the-art performance in terms of calibration, robustness and uncertainty characterisation over large-scale, multi-million parameter image classification tasks.

arXiv:2010.13317

Structural Prior Driven Regularized Deep Learning for Sonar Image Classification

Deep learning has been recently shown to improve performance in the domain of synthetic aperture sonar (SAS) image classification. Given the constant resolution with range of a SAS, it is no surprise that deep learning techniques perform so well. Despite deep learning's recent success, there are still compelling open challenges in reducing the high false alarm rate and enabling success when training imagery is limited, which is a practical challenge that distinguishes the SAS classification problem from standard image classification set-ups where training imagery may be abundant. We address these challenges by exploiting prior knowledge that humans use to grasp the scene. These include unconscious elimination of the image speckle and localization of objects in the scene. We introduce a new deep learning architecture which incorporates these priors with the goal of improving automatic target recognition (ATR) from SAS imagery. Our proposal -- called SPDRDL, Structural Prior Driven Regularized Deep Learning -- incorporates the previously mentioned priors in a multi-task convolutional neural network (CNN) and requires no additional training data when compared to traditional SAS ATR methods. Two structural priors are enforced via regularization terms in the learning of the network: (1) structural similarity prior -- enhanced imagery (often through despeckling) aids human interpretation and is semantically similar to the original imagery and (2) structural scene context priors -- learned features ideally encapsulate target centering information; hence learning may be enhanced via a regularization that encourages fidelity against known ground truth target shifts (relative target position from scene center). Experiments on a challenging real-world dataset reveal that SPDRDL outperforms state-of-the-art deep learning and other competing methods for SAS image classification.

arXiv:2010.13841

Peak Detection On Data Independent Acquisition Mass Spectrometry Data With Semisupervised Convolutional Transformers

Liquid Chromatography coupled to Mass Spectrometry (LC-MS) based methods are commonly used for high-throughput, quantitative measurements of the proteome (i.e. the set of all proteins in a sample at a given time). Targeted LC-MS produces data in the form of a two-dimensional time series spectrum, with the mass to charge ratio of analytes (m/z) on one axis, and the retention time from the chromatography on the other. The elution of a peptide of interest produces highly specific patterns across multiple fragment ion traces (extracted ion chromatograms, or XICs). In this paper, we formulate this peak detection problem as a multivariate time series segmentation problem, and propose a novel approach based on the Transformer architecture. Here we augment Transformers, which are capable of capturing long distance dependencies with a global view, with Convolutional Neural Networks (CNNs), which can capture local context important to the task at hand, in the form of Transformers with Convolutional Self-Attention. We further train this model in a semisupervised manner by adapting state of the art semisupervised image classification techniques for multi-channel time series data. Experiments on a representative LC-MS dataset are benchmarked using manual annotations to showcase the encouraging performance of our method; it outperforms baseline neural network architectures and is competitive against the current state of the art in automated peak detection.

arXiv:2010.13737공식 코드 · Papers with Code

Real-Time Edge Classification: Optimal Offloading under Token Bucket Constraints

To deploy machine learning-based algorithms for real-time applications with strict latency constraints, we consider an edge-computing setting where a subset of inputs are offloaded to the edge for processing by an accurate but resource-intensive model, and the rest are processed only by a less-accurate model on the device itself. Both models have computational costs that match available compute resources, and process inputs with low-latency. But offloading incurs network delays, and to manage these delays to meet application deadlines, we use a token bucket to constrain the average rate and burst length of transmissions from the device. We introduce a Markov Decision Process-based framework to make offload decisions under these constraints, based on the local model's confidence and the token bucket state, with the goal of minimizing a specified error measure for the application. Beyond isolated decisions for individual devices, we also propose approaches to allow multiple devices connected to the same access switch to share their bursting allocation. We evaluate and analyze the policies derived using our framework on the standard ImageNet image classification benchmark.

arXiv:2010.14246공식 코드 · Papers with Code

$μ$NAS: Constrained Neural Architecture Search for Microcontrollers

IoT devices are powered by microcontroller units (MCUs) which are extremely resource-scarce: a typical MCU may have an underpowered processor and around 64 KB of memory and persistent storage, which is orders of magnitude fewer computational resources than is typically required for deep learning. Designing neural networks for such a platform requires an intricate balance between keeping high predictive performance (accuracy) while achieving low memory and storage usage and inference latency. This is extremely challenging to achieve manually, so in this work, we build a neural architecture search (NAS) system, called $\mu$NAS, to automate the design of such small-yet-powerful MCU-level networks. $\mu$NAS explicitly targets the three primary aspects of resource scarcity of MCUs: the size of RAM, persistent storage and processor speed. $\mu$NAS represents a significant advance in resource-efficient models, especially for "mid-tier" MCUs with memory requirements ranging from 0.5 KB to 64 KB. We show that on a variety of image classification datasets $\mu$NAS is able to (a) improve top-1 classification accuracy by up to 4.8%, or (b) reduce memory footprint by 4--13x, or (c) reduce the number of multiply-accumulate operations by at least 2x, compared to existing MCU specialist literature and resource-efficient models.

arXiv:2010.14149

Active Learning for Noisy Data Streams Using Weak and Strong Labelers

Labeling data correctly is an expensive and challenging task in machine learning, especially for on-line data streams. Deep learning models especially require a large number of clean labeled data that is very difficult to acquire in real-world problems. Choosing useful data samples to label while minimizing the cost of labeling is crucial to maintain efficiency in the training process. When confronted with multiple labelers with different expertise and respective labeling costs, deciding which labeler to choose is nontrivial. In this paper, we consider a novel weak and strong labeler problem inspired by humans natural ability for labeling, in the presence of data streams with noisy labels and constrained by a limited budget. We propose an on-line active learning algorithm that consists of four steps: filtering, adding diversity, informative sample selection, and labeler selection. We aim to filter out the suspicious noisy samples and spend the budget on the diverse informative data using strong and weak labelers in a cost-effective manner. We derive a decision function that measures the information gain by combining the informativeness of individual samples and model confidence. We evaluate our proposed algorithm on the well-known image classification datasets CIFAR10 and CIFAR100 with up to 60% noise. Experiments show that by intelligently deciding which labeler to query, our algorithm maintains the same accuracy compared to the case of having only one of the labelers available while spending less of the budget.

arXiv:2010.13514공식 코드 · Papers with Code

Delta-STN: Efficient Bilevel Optimization for Neural Networks using Structured Response Jacobians

Hyperparameter optimization of neural networks can be elegantly formulated as a bilevel optimization problem. While research on bilevel optimization of neural networks has been dominated by implicit differentiation and unrolling, hypernetworks such as Self-Tuning Networks (STNs) have recently gained traction due to their ability to amortize the optimization of the inner objective. In this paper, we diagnose several subtle pathologies in the training of STNs. Based on these observations, we propose the $\Delta$-STN, an improved hypernetwork architecture which stabilizes training and optimizes hyperparameters much more efficiently than STNs. The key idea is to focus on accurately approximating the best-response Jacobian rather than the full best-response function; we achieve this by reparameterizing the hypernetwork and linearizing the network around the current parameters. We demonstrate empirically that our $\Delta$-STN can tune regularization hyperparameters (e.g. weight decay, dropout, number of cutout holes) with higher accuracy, faster convergence, and improved stability compared to existing approaches.

arXiv:2010.12575

Explanation and Use of Uncertainty Quantified by Bayesian Neural Network Classifiers for Breast Histopathology Images

Despite the promise of Convolutional neural network (CNN) based classification models for histopathological images, it is infeasible to quantify its uncertainties. Moreover, CNNs may suffer from overfitting when the data is biased. We show that Bayesian-CNN can overcome these limitations by regularizing automatically and by quantifying the uncertainty. We have developed a novel technique to utilize the uncertainties provided by the Bayesian-CNN that significantly improves the performance on a large fraction of the test data (about 6% improvement in accuracy on 77% of test data). Further, we provide a novel explanation for the uncertainty by projecting the data into a low dimensional space through a nonlinear dimensionality reduction technique. This dimensionality reduction enables interpretation of the test data through visualization and reveals the structure of the data in a low dimensional feature space. We show that the Bayesian-CNN can perform much better than the state-of-the-art transfer learning CNN (TL-CNN) by reducing the false negative and false positive by 11% and 7.7% respectively for the present data set. It achieves this performance with only 1.86 million parameters as compared to 134.33 million for TL-CNN. Besides, we modify the Bayesian-CNN by introducing a stochastic adaptive activation function. The modified Bayesian-CNN performs slightly better than Bayesian-CNN on all performance metrics and significantly reduces the number of false negatives and false positives (3% reduction for both). We also show that these results are statistically significant by performing McNemar's statistical significance test. This work shows the advantages of Bayesian-CNN against the state-of-the-art, explains and utilizes the uncertainties for histopathological images. It should find applications in various medical image classifications.

arXiv:2010.12337공식 코드 · Papers with Code

Fusion of Dual Spatial Information for Hyperspectral Image Classification

The inclusion of spatial information into spectral classifiers for fine-resolution hyperspectral imagery has led to significant improvements in terms of classification performance. The task of spectral-spatial hyperspectral image classification has remained challenging because of high intraclass spectrum variability and low interclass spectral variability. This fact has made the extraction of spatial information highly active. In this work, a novel hyperspectral image classification framework using the fusion of dual spatial information is proposed, in which the dual spatial information is built by both exploiting pre-processing feature extraction and post-processing spatial optimization. In the feature extraction stage, an adaptive texture smoothing method is proposed to construct the structural profile (SP), which makes it possible to precisely extract discriminative features from hyperspectral images. The SP extraction method is used here for the first time in the remote sensing community. Then, the extracted SP is fed into a spectral classifier. In the spatial optimization stage, a pixel-level classifier is used to obtain the class probability followed by an extended random walker-based spatial optimization technique. Finally, a decision fusion rule is utilized to fuse the class probabilities obtained by the two different stages. Experiments performed on three data sets from different scenes illustrate that the proposed method can outperform other state-of-the-art classification techniques. In addition, the proposed feature extraction method, i.e., SP, can effectively improve the discrimination between different land covers.

arXiv:2010.11697공식 코드 · Papers with Code

A Data Set and a Convolutional Model for Iconography Classification in Paintings

Iconography in art is the discipline that studies the visual content of artworks to determine their motifs and themes andto characterize the way these are represented. It is a subject of active research for a variety of purposes, including the interpretation of meaning, the investigation of the origin and diffusion in time and space of representations, and the study of influences across artists and art works. With the proliferation of digital archives of art images, the possibility arises of applying Computer Vision techniques to the analysis of art images at an unprecedented scale, which may support iconography research and education. In this paper we introduce a novel paintings data set for iconography classification and present the quantitativeand qualitative results of applying a Convolutional Neural Network (CNN) classifier to the recognition of the iconography of artworks. The proposed classifier achieves good performances (71.17% Precision, 70.89% Recall, 70.25% F1-Score and 72.73% Average Precision) in the task of identifying saints in Christian religious paintings, a task made difficult by the presence of classes with very similar visual features. Qualitative analysis of the results shows that the CNN focuses on the traditional iconic motifs that characterize the representation of each saint and exploits such hints to attain correct identification. The ultimate goal of our work is to enable the automatic extraction, decomposition, and comparison of iconography elements to support iconographic studies and automatic art work annotation.

arXiv:2010.11686

A Very Compact Embedded CNN Processor Design Based on Logarithmic Computing

In this paper, we propose a very compact embedded CNN processor design based on a modified logarithmic computing method using very low bit-width representation. Our high-quality CNN processor can easily fit into edge devices. For Yolov2, our processing circuit takes only 0.15 mm2 using TSMC 40 nm cell library. The key idea is to constrain the activation and weight values of all layers uniformly to be within the range [-1, 1] and produce low bit-width logarithmic representation. With the uniform representations, we devise a unified, reusable CNN computing kernel and significantly reduce computing resources. The proposed approach has been extensively evaluated on many popular image classification CNN models (AlexNet, VGG16, and ResNet-18/34) and object detection models (Yolov2). The hardware-implemented results show that our design consumes only minimal computing and storage resources, yet attains very high accuracy. The design is thoroughly verified on FPGAs, and the SoC integration is underway with promising results. With extremely efficient resource and energy usage, our design is excellent for edge computing purposes.

arXiv:2010.11692

Conversion and Implementation of State-of-the-Art Deep Learning Algorithms for the Classification of Diabetic Retinopathy

Diabetic retinopathy (DR) is a retinal microvascular condition that emerges in diabetic patients. DR will continue to be a leading cause of blindness worldwide, with a predicted 191.0 million globally diagnosed patients in 2030. Microaneurysms, hemorrhages, exudates, and cotton wool spots are common signs of DR. However, they can be small and hard for human eyes to detect. Early detection of DR is crucial for effective clinical treatment. Existing methods to classify images require much time for feature extraction and selection, and are limited in their performance. Convolutional Neural Networks (CNNs), as an emerging deep learning (DL) method, have proven their potential in image classification tasks. In this paper, comprehensive experimental studies of implementing state-of-the-art CNNs for the detection and classification of DR are conducted in order to determine the top performing classifiers for the task. Five CNN classifiers, namely Inception-V3, VGG19, VGG16, ResNet50, and InceptionResNetV2, are evaluated through experiments. They categorize medical images into five different classes based on DR severity. Data augmentation and transfer learning techniques are applied since annotated medical images are limited and imbalanced. Experimental results indicate that the ResNet50 classifier has top performance for binary classification and that the InceptionResNetV2 classifier has top performance for multi-class DR classification.

arXiv:2010.11682

Lung Nodule Classification Using Biomarkers, Volumetric Radiomics and 3D CNNs

We present a hybrid algorithm to estimate lung nodule malignancy that combines imaging biomarkers from Radiologist's annotation with image classification of CT scans. Our algorithm employs a 3D Convolutional Neural Network (CNN) as well as a Random Forest in order to combine CT imagery with biomarker annotation and volumetric radiomic features. We analyze and compare the performance of the algorithm using only imagery, only biomarkers, combined imagery + biomarkers, combined imagery + volumetric radiomic features and finally the combination of imagery + biomarkers + volumetric features in order to classify the suspicion level of nodule malignancy. The National Cancer Institute (NCI) Lung Image Database Consortium (LIDC) IDRI dataset is used to train and evaluate the classification task. We show that the incorporation of semi-supervised learning by means of K-Nearest-Neighbors (KNN) can increase the available training sample size of the LIDC-IDRI thereby further improving the accuracy of malignancy estimation of most of the models tested although there is no significant improvement with the use of KNN semi-supervised learning if image classification with CNNs and volumetric features are combined with descriptive biomarkers. Unexpectedly, we also show that a model using image biomarkers alone is more accurate than one that combines biomarkers with volumetric radiomics, 3D CNNs, and semi-supervised learning. We discuss the possibility that this result may be influenced by cognitive bias in LIDC-IDRI because malignancy estimates were recorded by the same radiologist panel as biomarkers, as well as future work to incorporate pathology information over a subset of study participants.

arXiv:2010.11426

Efficient Scale-Permuted Backbone with Learned Resource Distribution

Recently, SpineNet has demonstrated promising results on object detection and image classification over ResNet model. However, it is unclear if the improvement adds up when combining scale-permuted backbone with advanced efficient operations and compound scaling. Furthermore, SpineNet is built with a uniform resource distribution over operations. While this strategy seems to be prevalent for scale-decreased models, it may not be an optimal design for scale-permuted models. In this work, we propose a simple technique to combine efficient operations and compound scaling with a previously learned scale-permuted architecture. We demonstrate the efficiency of scale-permuted model can be further improved by learning a resource distribution over the entire network. The resulting efficient scale-permuted models outperform state-of-the-art EfficientNet-based models on object detection and achieve competitive performance on image classification and semantic segmentation. Code and models will be open-sourced soon.

arXiv:2010.11422

Learning Loss for Test-Time Augmentation

Data augmentation has been actively studied for robust neural networks. Most of the recent data augmentation methods focus on augmenting datasets during the training phase. At the testing phase, simple transformations are still widely used for test-time augmentation. This paper proposes a novel instance-level test-time augmentation that efficiently selects suitable transformations for a test input. Our proposed method involves an auxiliary module to predict the loss of each possible transformation given the input. Then, the transformations having lower predicted losses are applied to the input. The network obtains the results by averaging the prediction results of augmented inputs. Experimental results on several image classification benchmarks show that the proposed instance-aware test-time augmentation improves the model's robustness against various corruptions.

arXiv:2010.11714공식 코드 · Papers with Code

Restoring Negative Information in Few-Shot Object Detection

Few-shot learning has recently emerged as a new challenge in the deep learning field: unlike conventional methods that train the deep neural networks (DNNs) with a large number of labeled data, it asks for the generalization of DNNs on new classes with few annotated samples. Recent advances in few-shot learning mainly focus on image classification while in this paper we focus on object detection. The initial explorations in few-shot object detection tend to simulate a classification scenario by using the positive proposals in images with respect to certain object class while discarding the negative proposals of that class. Negatives, especially hard negatives, however, are essential to the embedding space learning in few-shot object detection. In this paper, we restore the negative information in few-shot object detection by introducing a new negative- and positive-representative based metric learning framework and a new inference scheme with negative and positive representatives. We build our work on a recent few-shot pipeline RepMet with several new modules to encode negative information for both training and testing. Extensive experiments on ImageNet-LOC and PASCAL VOC show our method substantially improves the state-of-the-art few-shot object detection solutions. Our code is available at https://github.com/yang-yk/NP-RepMet.

arXiv:2010.11521

Malaria detection from RBC images using shallow Convolutional Neural Networks

The advent of Deep Learning models like VGG-16 and Resnet-50 has considerably revolutionized the field of image classification, and by using these Convolutional Neural Networks (CNN) architectures, one can get a high classification accuracy on a wide variety of image datasets. However, these Deep Learning models have a very high computational complexity and so incur a high computational cost of running these algorithms as well as make it hard to interpret the results. In this paper, we present a shallow CNN architecture which gives the same classification accuracy as the VGG-16 and Resnet-50 models for thin blood smear RBC slide images for detection of malaria, while decreasing the computational run time by an order of magnitude. This can offer a significant advantage for commercial deployment of these algorithms, especially in poorer countries in Africa and some parts of the Indian subcontinent, where the menace of malaria is quite severe.