Conditioning analysis uncovers the landscape of an optimization objective by exploring the spectrum of its curvature matrix. This has been well explored theoretically for linear models. We extend this analysis to deep neural networks (DNNs) in order to investigate their learning dynamics. To this end, we propose layer-wise conditioning analysis, which explores the optimization landscape with respect to each layer independently. Such an analysis is theoretically supported under mild assumptions that approximately hold in practice. Based on our analysis, we show that batch normalization (BN) can stabilize the training, but sometimes result in the false impression of a local minimum, which has detrimental effects on the learning. Besides, we experimentally observe that BN can improve the layer-wise conditioning of the optimization problem. Finally, we find that the last linear layer of a very deep residual network displays ill-conditioned behavior. We solve this problem by only adding one BN layer before the last linear layer, which achieves improved performance over the original and pre-activation residual networks.
Weight pruning methods of DNNs have been demonstrated to achieve a good model pruning rate without loss of accuracy, thereby alleviating the significant computation/storage requirements of large-scale DNNs. Structured weight pruning methods have been proposed to overcome the limitation of irregular network structure and demonstrated actual GPU acceleration. However, in prior work the pruning rate (degree of sparsity) and GPU acceleration are limited (to less than 50%) when accuracy needs to be maintained. In this work,we overcome these limitations by proposing a unified, systematic framework o…
Huazhong University of Science and Technology（华中科技大学）；Baidu Research（百度研究院）；Nanyang Technological University（南洋理工大学）
Unlike the conventional wisdom in statistical learning theory, the test error of a deep neural network (DNN) often demonstrates double descent: as the model complexity increases, it first follows a classical U-shaped curve and then shows a second descent. Through bias-variance decomposition, recent studies revealed that the bell-shaped variance is the major cause of model-wise double descent (when the DNN is widened gradually). This paper investigates epoch-wise double descent, i.e., the test error of a DNN also shows double descent as the number of training epoches increases. By extending the…
The Hong Kong University of Science and Technology（香港科技大学）；Hong Kong Baptist University（香港浸会大学）
Distributed synchronous stochastic gradient descent has been widely used to train deep neural networks (DNNs) on computer clusters. With the increase of computational power, network communications generally limit the system scalability. Wait-free backpropagation (WFBP) is a popular solution to overlap communications with computations during the training process. In this paper, we observe that many DNNs have a large number of layers with only a small amount of data to be communicated at each layer in distributed training, which could make WFBP inefficient. Based on the fact that merging some sh…
Southern University of Science and Technology（南方科技大学）；Shanghai Jiao Tong University（上海交通大学）
This paper presents a method to explain how the information of each input variable is gradually discarded during the forward propagation in a deep neural network (DNN), which provides new perspectives to explain DNNs. We define two types of entropy-based metrics, i.e. (1) the discarding of pixel-wise information used in the forward propagation, and (2) the uncertainty of the input reconstruction, to measure input information contained by a specific layer from two perspectives. Unlike previous attribution metrics, the proposed metrics ensure the fairness of comparisons between different layers …
Generative Adversarial Networks (GANs) have proven successful for unsupervised image generation. Several works extended GANs to image inpainting by conditioning the generation with parts of the image one wants to reconstruct. However, these methods have limitations in settings where only a small subset of the image pixels is known beforehand. In this paper, we study the effectiveness of conditioning GANs by adding an explicit regularization term to enforce pixel-wise conditions when very few pixel values are provided. In addition, we also investigate the influence of this regularization term o…
The state-of-the-art deep neural networks (DNNs) have significant computational and data management requirements. The size of both training data and models continue to increase. Sparsification and pruning methods are shown to be effective in removing a large fraction of connections in DNNs. The resulting sparse networks present unique challenges to further improve the computational efficiency of training and inference in deep learning. Both the feedforward (inference) and backpropagation steps in stochastic gradient descent (SGD) algorithm for training sparse DNNs involve consecutive sparse ma…
University of Rhode Island（罗德岛大学）；Northeastern University（东北大学）
Feature selection is central to contemporary high-dimensional data analysis. Grouping structure among features arises naturally in various scientific problems. Many methods have been proposed to incorporate the grouping structure information into feature selection. However, these methods are normally restricted to a linear regression setting. To relax the linear constraint, we combine the deep neural networks (DNNs) with the recent Knockoffs technique, which has been successful in an individual feature selection context. We propose Deep-gKnock (Deep group-feature selection using Knockoffs) as …
Mix-and-Match Pruning: Globally Guided Layer-Wise Sparsification of DNNs
1 Brandenburg Technical University, Cottbus, Germany；2 University of Zanjan, Iran；3 Tallinn University of Technology, Estonia；4 Leibniz Institute for High Performance Microelectronics, Frankfurt Oder, Germany
Fudan University（复旦大学）；Deakin University, Geelong（迪肯大学）；Peking University（北京大学）；The University of Melbourne（墨尔本大学）；University of Illinois at Urbana–Champaign（伊利诺伊大学厄巴纳-香槟分校）
Deep neural networks (DNNs) are increasingly deployed in different applications to achieve state-of-the-art performance. However, they are often applied as a black box with limited understanding of what knowledge the model has learned from the data. In this paper, we focus on image classification and propose a method to visualize and understand the class-wise knowledge (patterns) learned by DNNs under three different settings including natural, backdoor and adversarial. Different to existing visualization methods, our method searches for a single predictive pattern in the pixel space to repres…
Generative Adversarial Networks (GANs) have proven successful for unsupervised image generation. Several works have extended GANs to image inpainting by conditioning the generation with parts of the image to be reconstructed. Despite their success, these methods have limitations in settings where only a small subset of the image pixels is known beforehand. In this paper we investigate the effectiveness of conditioning GANs when very few pixel values are provided. We propose a modelling framework which results in adding an explicit cost term to the GAN objective function to enforce pixel-wise c…
The University of Electronic Science and Technology of China（电子科技大学）
Although great progress has been made on adversarial attacks for deep neural networks (DNNs), their transferability is still unsatisfactory, especially for targeted attacks. There are two problems behind that have been long overlooked: 1) the conventional setting of $T$ iterations with the step size of $ε/T$ to comply with the $ε$-constraint. In this case, most of the pixels are allowed to add very small noise, much less than $ε$; and 2) usually manipulating pixel-wise noise. However, features of a pixel extracted by DNNs are influenced by its surrounding regions, and different DNNs generally …
Inception Institute of Artificial Intelligence（人工智能起源研究所）；University of Electronic Science and Technology of China（电子科技大学）；ByteDance AI Lab（字节跳动人工智能实验室）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Orthogonality is widely used for training deep neural networks (DNNs) due to its ability to maintain all singular values of the Jacobian close to 1 and reduce redundancy in representation. This paper proposes a computationally efficient and numerically stable orthogonalization method using Newton's iteration (ONI), to learn a layer-wise orthogonal weight matrix in DNNs. ONI works by iteratively stretching the singular values of a weight matrix towards 1. This property enables it to control the orthogonality of a weight matrix by its number of iterations. We show that our method improves the pe…
Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization
Alma Mater Studiorum - University of Bologna（博洛尼亚大学）；Center for Biomolecular Nanotechnologies, Istituto Italiano di Tecnologia（意大利技术研究院生物分子纳米技术中心）；School of Medical Sciences, University of Campinas (UNICAMP)（坎皮纳斯大学（UNICAMP）医学院）；Istituto Italiano di Tecnologia（意大利技术研究院）
University of Electronic Science and Technology of China（电子科技大学）；Beihang University（北京航空航天大学）
By adding human-imperceptible noise to clean images, the resultant adversarial examples can fool other unknown models. Features of a pixel extracted by deep neural networks (DNNs) are influenced by its surrounding regions, and different DNNs generally focus on different discriminative regions in recognition. Motivated by this, we propose a patch-wise iterative algorithm -- a black-box attack towards mainstream normally trained and defense models, which differs from the existing attack methods manipulating pixel-wise noise. In this way, without sacrificing the performance of white-box attack, o…
LILAC: Layer-Wise Independent LoRAs and Cascaded Conditioning for Multi-Concept Customization of Diffusion Models
Adobe Research, Romania（罗马尼亚Adobe研究院）；Department of Computer Science, University of Bucharest, Romania（罗马尼亚布加勒斯特大学计算机科学系）；International Computer High School of Bucharest, Romania（罗马尼亚布加勒斯特国际计算机高中）
South China University of Technology（华南理工大学）；The University of Sydney（悉尼大学）
Many machine learning problems concern with discovering or associating common patterns in data of multiple views or modalities. Multi-view learning is of the methods to achieve such goals. Recent methods propose deep multi-view networks via adaptation of generic Deep Neural Networks (DNNs), which concatenate features of individual views at intermediate network layers (i.e., fusion layers). In this work, we study the problem of multi-view learning in such end-to-end networks. We take a regularization approach via multi-view learning criteria, and propose a novel, effective, and efficient neuron…
Purdue University（普渡大学）；IBM T.J. Watson Research Center（IBM T.J. 沃森研究中心）
Precision scaling has emerged as a popular technique to optimize the compute and storage requirements of Deep Neural Networks (DNNs). Efforts toward creating ultra-low-precision (sub-8-bit) DNNs suggest that the minimum precision required to achieve a given network-level accuracy varies considerably across networks, and even across layers within a network, requiring support for variable precision in DNN hardware. Previous proposals such as bit-serial hardware incur high overheads, significantly diminishing the benefits of lower precision. To efficiently support precision re-configurability in …
MILA, Université de Montréal（蒙特利尔大学MILA）；Rice University（莱斯大学）；Univ. Lille, CNRS, Centrale Lille, Inria, UMR 9189 CRIStAL（里尔大学、法国国家科学研究中心、里尔中央理工学院、法国国家信息与自动化研究所、UMR 9189 CRIStAL实验室）；CIFAR Fellow（加拿大高等研究院研究员）
We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning - answering image-related questions which require a multi-step, high-level process - a task which has proven difficult for standard deep learning methods that do not explicitly model reasoning. Specifically, we show on visual reasoning tasks that FiLM layers 1) halve state-of-the-art…
Semantic Routing: Exploring Multi-Layer LLM Feature Weighting for Diffusion Transformers
Peking University（北京大学）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Nanjing University（南京大学）；Fudan University（复旦大学）
Neural Field Conditioning Strategies for 2D Semantic Segmentation
University of Hamburg（汉堡大学）；ZAL Center of Applied Aeronautical Research（ZAL应用航空研究中心）；Hamburger Informatik Technologie-Center e.V. (HITeC)（汉堡信息技术中心（HITeC））
Fisher Vector classifiers and Deep Neural Networks (DNNs) are popular and successful algorithms for solving image classification problems. However, both are generally considered `black box' predictors as the non-linear transformations involved have so far prevented transparent and interpretable reasoning. Recently, a principled technique, Layer-wise Relevance Propagation (LRP), has been developed in order to better comprehend the inherent structured reasoning of complex nonlinear classification models such as Bag of Feature models or DNNs. In this paper we (1) extend the LRP framework also for…
It is known that Deep Neural networks (DNNs) are vulnerable to adversarial attacks, and the adversarial robustness of DNNs could be improved by adding adversarial noises to training data (e.g., the standard adversarial training (SAT)). However, inappropriate noises added to training data may reduce a model's performance, which is termed the trade-off between accuracy and robustness. This problem has been sufficiently studied for the classification of whole images but has rarely been explored for image analysis tasks in the medical application domain, including image segmentation, landmark dete…
University of Lübeck（吕贝克大学）；Robert Bosch GmbH（罗伯特·博世有限公司）
Deep neural networks (DNNs) are vulnerable to adversarial examples and other data perturbations. Especially in safety critical applications of DNNs, it is therefore crucial to detect misclassified samples. The current state-of-the-art detection methods require either significantly more runtime or more parameters than the original network itself. This paper therefore proposes GraN, a time- and parameter-efficient method that is easily adaptable to any DNN. GraN is based on the layer-wise norm of the DNN's gradient regarding the loss of the current input-output combination, which can be computed…
In this paper, we propose training very deep neural networks (DNNs) for supervised learning of hash codes. Existing methods in this context train relatively "shallow" networks limited by the issues arising in back propagation (e.e. vanishing gradients) as well as computational efficiency. We propose a novel and efficient training algorithm inspired by alternating direction method of multipliers (ADMM) that overcomes some of these limitations. Our method decomposes the training process into independent layer-wise local updates through auxiliary variables. Empirically we observe that our trainin…
Classifiers that are linear in their parameters, and trained by optimizing a convex loss function, have predictable behavior with respect to changes in the training data, initial conditions, and optimization. Such desirable properties are absent in deep neural networks (DNNs), typically trained by non-linear fine-tuning of a pre-trained model. Previous attempts to linearize DNNs have led to interesting theoretical insights, but have not impacted the practice due to the substantial performance gap compared to standard non-linear optimization. We present the first method for linearizing a pre-tr…
University of Chicago（芝加哥大学）；Princeton University（普林斯顿大学）
Representation learning, i.e. the generation of representations useful for downstream applications, is a task of fundamental importance that underlies much of the success of deep neural networks (DNNs). Recently, robustness to adversarial examples has emerged as a desirable property for DNNs, spurring the development of robust training methods that account for adversarial examples. In this paper, we aim to understand how the properties of representations learned by robust training differ from those obtained from standard, non-robust training. This is critical to diagnosing numerous salient pit…
Polytechnique Montreal（蒙特利尔理工学院）；Mila, Quebec AI Institute（魁北克人工智能研究所米拉）；Beijing Tiantan Hospital, Capital Medical University（首都医科大学附属北京天坛医院）；Stanford University Center for Artificial Intelligence in Medicine & Imaging（斯坦福大学医学与影像人工智能中心）；University of Montreal（蒙特利尔大学）
Medical images are often accompanied by metadata describing the image (vendor, acquisition parameters) and the patient (disease type or severity, demographics, genomics). This metadata is usually disregarded by image segmentation methods. In this work, we adapt a linear conditioning method called FiLM (Feature-wise Linear Modulation) for image segmentation tasks. This FiLM adaptation enables integrating metadata into segmentation models for better performance. We observed an average Dice score increase of 5.1% on spinal cord tumor segmentation when incorporating the tumor type with FiLM. The m…
Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute, HHI（弗劳恩霍夫电信研究所，海因里希·赫兹研究所，HHI）；Humboldt University Berlin（柏林洪堡大学）
The new era of image segmentation leveraging the power of Deep Neural Nets (DNNs) comes with a price tag: to train a neural network for pixel-wise segmentation, a large amount of training samples has to be manually labeled on pixel-precision. In this work, we address this by following an indirect solution. We build upon the advances of the Explainable AI (XAI) community and extract a pixel-wise binary segmentation from the output of the Layer-wise Relevance Propagation (LRP) explaining the decision of a classification network. We show that we achieve similar results compared to an established …
Medical University of Vienna（维也纳医科大学）；Technical University Graz（格拉茨工业大学）；Complexity Science Hub（复杂性科学中心）
Recently, pruning deep neural networks (DNNs) has received a lot of attention for improving accuracy and generalization power, reducing network size, and increasing inference speed on specialized hardwares. Although pruning was mainly tested on computer vision tasks, its application in the context of medical image analysis has hardly been explored. This work investigates the impact of well-known pruning techniques, namely layer-wise and network-wide magnitude pruning, on the nuclei instance segmentation performance in histological images. Our utilized instance segmentation model consists of tw…
