本文提出 RNAS-CL，利用跨层知识蒸馏让 NAS 自动匹配教师层监督学生层，从而搜索出小型且对对抗攻击鲁棒的神经架构。
摘要：
深度神经网络容易受到对抗攻击。神经架构搜索（NAS）作为深度神经网络的核心推动工具之一，在多种机器学习应用中展现出优异的预测准确率。然而，它在对抗攻击下的表现仍不明确。在存在鲁棒教师模型的情况下，研究 NAS 是否能够通过继承教师模型的鲁棒性来生成鲁棒的神经架构，将是一个有意义的问题。本文提出 Robust Neural Architecture Search by Cross-Layer Knowledge Distillation（RNAS-CL），这是一种新颖的 NAS 算法，通过跨层知识蒸馏向鲁棒教师模型学习，以提升 NAS 的鲁棒性。以往的知识蒸馏方法只鼓励学生模型与教师模型在最后一层的输出接近，与之不同，RNAS-CL 会自动搜索最佳的教师层来监督每个学生层。实验结果证明了 RNAS-CL 的有效性，并表明 RNAS-CL 能够生成小型且鲁棒的神经架构。
英文摘要：
Deep Neural Networks are vulnerable to adversarial attacks. Neural Architecture Search (NAS), one of the driving tools of deep neural networks, demonstrates superior performance in prediction accuracy in various machine learning applications. However, it is unclear how it performs against adversarial attacks. Given the presence of a robust teacher, it would be interesting to investigate if NAS would produce robust neural architecture by inheriting robustness from the teacher. In this paper, we propose Robust Neural Architecture Search by Cross-Layer Knowledge Distillation (RNAS-CL), a novel NAS algorithm that improves the robustness of NAS by learning from a robust teacher through cross-layer knowledge distillation. Unlike previous knowledge distillation methods that encourage close student/teacher output only in the last layer, RNAS-CL automatically searches for the best teacher layer to supervise each student layer. Experimental result evidences the effectiveness of RNAS-CL and shows that RNAS-CL produces small and robust neural architecture.
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）
Peking University（北京大学）；Megvii (Face++) Technology Inc（旷视科技（Face++））；National Engineering Laboratory for Big Data Analysis and Applications（大数据分析与应用技术国家工程实验室）；DiDi AI Labs（滴滴人工智能实验室）
Knowledge distillation has become increasingly important in model compression. It boosts the performance of a miniaturized student network with the supervision of the output distribution and feature maps from a sophisticated teacher network. Some recent works introduce multi-teacher distillation to provide more supervision to the student network. However, the effectiveness of multi-teacher distillation methods are accompanied by costly computation resources. To tackle with both the efficiency and the effectiveness of knowledge distillation, we introduce the feature aggregation to imitate the m…
Erasmus Medical Center（伊拉斯姆斯医学中心）；IBM Research（IBM研究院）
RNA-binding proteins (RBPs) play crucial roles in many biological processes, e.g. gene regulation. Computational identification of RBP binding sites on RNAs are urgently needed. In particular, RBPs bind to RNAs by recognizing sequence motifs. Thus, fast locating those motifs on RNA sequences is crucial and time-efficient for determining whether the RNAs interact with the RBPs or not. In this study, we present an attention based convolutional neural network, iDeepA, to predict RNA-protein binding sites from raw RNA sequences. We first encode RNA sequences into one-hot encoding. Next, we design …
Istanbul Technical University（伊斯坦布尔理工大学）；University of Maryland, Baltimore County（马里兰大学巴尔的摩县分校）；Arcelik Research and Development（阿奇立克研发中心）
One of the most efficient methods for model compression is hint distillation, where the student model is injected with information (hints) from several different layers of the teacher model. Although the selection of hint points can drastically alter the compression performance, conventional distillation approaches overlook this fact and use the same hint points as in the early studies. Therefore, we propose a clustering based hint selection methodology, where the layers of teacher model are clustered with respect to several metrics and the cluster centers are used as the hint points. Our meth…
National University of Defense Technology（国防科技大学）；Chinese Academy of Sciences（中国科学院）
Knowledge distillation (KD) is an effective training strategy to improve the lightweight student models under the guidance of cumbersome teachers. However, the large architecture difference across the teacher-student pairs limits the distillation gains. In contrast to previous adaptive distillation methods to reduce the teacher-student gap, we explore a novel training-free framework to search for the best student architectures for a given teacher. Our work first empirically show that the optimal model under vanilla training cannot be the winner in distillation. Secondly, we find that the simil…
We present a novel framework of knowledge distillation that is capable of learning powerful and efficient student models from ensemble teacher networks. Our approach addresses the inherent model capacity issue between teacher and student and aims to maximize benefit from teacher models during distillation by reducing their capacity gap. Specifically, we employ a neural architecture search technique to augment useful structures and operations, where the searched network is appropriate for knowledge distillation towards student models and free from sacrificing its performance by fixing the netwo…
Recent Knowledge distillation (KD) studies show that different manually designed schemes impact the learned results significantly. Yet, in KD, automatically searching an optimal distillation scheme has not yet been well explored. In this paper, we propose DistPro, a novel framework which searches for an optimal KD process via differentiable meta-learning. Specifically, given a pair of student and teacher networks, DistPro first sets up a rich set of KD connection from the transmitting layers of the teacher to the receiving layers of the student, and in the meanwhile, various transforms are als…
Improving Differentiable Architecture Search via Self-Distillation
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）；Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）
We introduce knowledge distillation for end-to-end person search. End-to-End methods are the current state-of-the-art for person search that solve both detection and re-identification jointly. These approaches for joint optimization show their largest drop in performance due to a sub-optimal detector. We propose two distinct approaches for extra supervision of end-to-end person search methods in a teacher-student setting. The first is adopted from state-of-the-art knowledge distillation in object detection. We employ this to supervise the detector of our person search model at various levels u…
In recent years, deeper and wider neural networks have shown excellent performance in computer vision tasks, while their enormous amount of parameters results in increased computational cost and overfitting. Several methods have been proposed to compress the size of the networks without reducing network performance. Network pruning can reduce redundant and unnecessary parameters from a network. Knowledge distillation can transfer the knowledge of deeper and wider networks to smaller networks. The performance of the smaller network obtained by these methods is bounded by the predefined network.…
This paper proposes a knowledge distillation method for foreground object search (FoS). Given a background and a rectangle specifying the foreground location and scale, FoS retrieves compatible foregrounds in a certain category for later image composition. Foregrounds within the same category can be grouped into a small number of patterns. Instances within each pattern are compatible with any query input interchangeably. These instances are referred to as interchangeable foregrounds. We first present a pipeline to build pattern-level FoS dataset containing labels of interchangeable foregrounds…
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究所）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）
Neural architecture search (NAS) targets at finding the optimal architecture of a neural network for a problem or a family of problems. Evaluations of neural architectures are very time-consuming. One of the possible ways to mitigate this issue is to use low-fidelity evaluations, namely training on a part of a dataset, fewer epochs, with fewer channels, etc. In this paper, we propose a bayesian multi-fidelity method for neural architecture search: MF-KD. The method relies on a new approach to low-fidelity evaluations of neural architectures by training for a few epochs using a knowledge distil…
Southeast University（东南大学）；Southeast University Shenzhen Research Institute（东南大学深圳研究院）
Nowadays, the rapid development of photovoltaic(PV) power stations requires increasingly reliable maintenance and fault diagnosis of PV modules in the field. Due to the effectiveness, convolutional neural network (CNN) has been widely used in the existing automatic defect detection of PV cells. However, the parameters of these CNN-based models are very large, which require stringent hardware resources and it is difficult to be applied in actual industrial projects. To solve these problems, we propose a novel lightweight high-performance model for automatic defect detection of PV cells in elect…
Tongji University（同济大学）；The University of Adelaide（阿德莱德大学）；Monash University（莫纳什大学）
Person search aims to localize and identify a specific person from a gallery of images. Recent methods can be categorized into two groups, i.e., two-step and end-to-end approaches. The former views person search as two independent tasks and achieves dominant results using separately trained person detection and re-identification (Re-ID) models. The latter performs person search in an end-to-end fashion. Although the end-to-end approaches yield higher inference efficiency, they largely lag behind those two-step counterparts in terms of accuracy. In this paper, we argue that the gap between the …
Optimized Federated Knowledge Distillation with Distributed Neural Architecture Search
Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg（安全、可靠与信任跨学科研究中心（SnT），卢森堡大学）；Blekinge Institute of Technology（布莱金厄理工大学）；ADSLabs, Umea University（ADSLabs，乌梅亚大学）
Nagoya University（名古屋大学）；Navier, Inc.（Navier公司）；The University of Tokyo（东京大学）；Ritsumeikan University（立命馆大学）；Hitotsubashi University（一桥大学）
The number of traffic accidents has been continuously increasing in recent years worldwide. Many accidents are caused by distracted drivers, who take their attention away from driving. Motivated by the success of Convolutional Neural Networks (CNNs) in computer vision, many researchers developed CNN-based algorithms to recognize distracted driving from a dashcam and warn the driver against unsafe behaviors. However, current models have too many parameters, which is unfeasible for vehicle-mounted computing. This work proposes a novel knowledge-distillation-based framework to solve this problem.…
Fraunhofer Institute for Computer Graphics Research IGD（弗劳恩霍夫计算机图形研究所）；Mathematical and Applied Visual Computing, TU Darmstadt（达姆施塔特工业大学数学与应用视觉计算）
Deep neural networks have rapidly become the mainstream method for face recognition (FR). However, this limits the deployment of such models that contain an extremely large number of parameters to embedded and low-end devices. In this work, we present an extremely lightweight and accurate FR solution, namely PocketNet. We utilize neural architecture search to develop a new family of lightweight face-specific architectures. We additionally propose a novel training paradigm based on knowledge distillation (KD), the multi-step KD, where the knowledge is distilled from the teacher model to the stu…
In many practical scenarios -- like hyperparameter search or continual retraining with new data -- related training runs are performed many times in sequence. Current practice is to train each of these models independently from scratch. We study the problem of exploiting the computation invested in previous runs to reduce the cost of future runs using knowledge distillation (KD). We find that augmenting future runs with KD from previous runs dramatically reduces the time necessary to train these models, even taking into account the overhead of KD. We improve on these results with two strategie…
DarkMatter AI Research（暗物质人工智能研究院）；Monash University（莫纳什大学）；Sun Yat-sen University（中山大学）
Neural Architecture Search (NAS), aiming at automatically designing network architectures by machines, is hoped and expected to bring about a new revolution in machine learning. Despite these high expectation, the effectiveness and efficiency of existing NAS solutions are unclear, with some recent works going so far as to suggest that many existing NAS solutions are no better than random architecture selection. The inefficiency of NAS solutions may be attributed to inaccurate architecture evaluation. Specifically, to speed up NAS, recent works have proposed under-training different candidate a…
SenseTime Research（商汤科技研究院）；The Chinese University of Hong Kong（香港中文大学）；Centre for Perceptual and Interactive Intelligence（感知与交互智能研究中心）
Recent studies pointed out that knowledge distillation (KD) suffers from two degradation problems, the teacher-student gap and the incompatibility with strong data augmentations, making it not applicable to training state-of-the-art models, which are trained with advanced augmentations. However, we observe that a key factor, i.e., the temperatures in the softmax functions for generating probabilities of both the teacher and student models, was mostly overlooked in previous methods. With properly tuned temperatures, such degradation problems of KD can be much mitigated. However, instead of rely…
Hong Kong University of Science and Technology（香港科技大学）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）
We propose Joint-DetNAS, a unified NAS framework for object detection, which integrates 3 key components: Neural Architecture Search, pruning, and Knowledge Distillation. Instead of naively pipelining these techniques, our Joint-DetNAS optimizes them jointly. The algorithm consists of two core processes: student morphism optimizes the student's architecture and removes the redundant parameters, while dynamic distillation aims to find the optimal matching teacher. For student morphism, weight inheritance strategy is adopted, allowing the student to flexibly update its architecture while fully u…
Google AI（谷歌人工智能）；Google Brain（谷歌大脑）；Multimedia Laboratory, The Chinese University of Hong Kong（香港中文大学多媒体实验室）
Standard Knowledge Distillation (KD) approaches distill the knowledge of a cumbersome teacher model into the parameters of a student model with a pre-defined architecture. However, the knowledge of a neural network, which is represented by the network's output distribution conditioned on its input, depends not only on its parameters but also on its architecture. Hence, a more generalized approach for KD is to distill the teacher's knowledge into both the parameters and architecture of the student. To achieve this, we present a new Architecture-aware Knowledge Distillation (AKD) approach that f…
Differential Neural Architecture Search (NAS) requires all layer choices to be held in memory simultaneously; this limits the size of both search space and final architecture. In contrast, Probabilistic NAS, such as PARSEC, learns a distribution over high-performing architectures, and uses only as much memory as needed to train a single model. Nevertheless, it needs to sample many architectures, making it computationally expensive for searching in an extensive space. To solve these problems, we propose a sampling method adaptive to the distribution entropy, drawing more samples to encourage ex…
Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区）；University of Oxford（牛津大学）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）；DarkMatter AI Research（暗物质人工智能研究院）
Pre-trained language models have shown remarkable results on various NLP tasks. Nevertheless, due to their bulky size and slow inference speed, it is hard to deploy them on edge devices. In this paper, we have a critical insight that improving the feed-forward network (FFN) in BERT has a higher gain than improving the multi-head attention (MHA) since the computational cost of FFN is 2$\sim$3 times larger than MHA. Hence, to compact BERT, we are devoted to designing efficient FFN as opposed to previous works that pay attention to MHA. Since FFN comprises a multilayer perceptron (MLP) that is es…
Centre for Frontier AI Research (CFAR), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局前沿人工智能研究中心（CFAR））；National University of Singapore（新加坡国立大学）；Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局高性能计算研究所（IHPC））；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Model-based deep learning has achieved astounding successes due in part to the availability of large-scale real-world data. However, processing such massive amounts of data comes at a considerable cost in terms of computations, storage, training and the search for good neural architectures. Dataset distillation has thus recently come to the fore. This paradigm involves distilling information from large real-world datasets into tiny and compact synthetic datasets such that processing the latter ideally yields similar performances as the former. State-of-the-art methods primarily rely on learnin…
Bit-mask Robust Contrastive Knowledge Distillation for Unsupervised Semantic Hashing
School of Computer Science and Technology, University of Science and Technology of China（中国科学技术大学计算机科学与技术学院）；School of Data Science, University of Science and Technology of China（中国科学技术大学数据科学学院）；iFLYTEK AI Research（讯飞AI研究院）
City University of Hong Kong（香港城市大学）；Tsinghua University（清华大学）；Microsoft Research Asia（微软亚洲研究院）；Chinese Academy of Sciences（中国科学院）
One-shot weight sharing methods have recently drawn great attention in neural architecture search due to high efficiency and competitive performance. However, weight sharing across models has an inherent deficiency, i.e., insufficient training of subnetworks in hypernetworks. To alleviate this problem, we present a simple yet effective architecture distillation method. The central idea is that subnetworks can learn collaboratively and teach each other throughout the training process, aiming to boost the convergence of individual models. We introduce the concept of prioritized path, which refer…
School of Systems Science, Beijing Normal University（北京师范大学系统科学学院）；School of Automation and Intelligent Manufacturing, Southern University of Science and Technology（南方科技大学自动化与智能制造学院）；Department of Electrical and Computer Engineering, University of Illinois Chicago（伊利诺伊大学芝加哥分校电气与计算机工程系）
Most existing distillation methods ignore the flexible role of the temperature in the loss function and fix it as a hyper-parameter that can be decided by an inefficient grid search. In general, the temperature controls the discrepancy between two distributions and can faithfully determine the difficulty level of the distillation task. Keeping a constant temperature, i.e., a fixed level of task difficulty, is usually sub-optimal for a growing student during its progressive learning stages. In this paper, we propose a simple curriculum-based technique, termed Curriculum Temperature for Knowledg…
National University of Defense Technology（国防科技大学）；Chinese Academy of Sciences（中国科学院）
Neural architecture search (NAS) has made tremendous progress in the automatic design of effective neural network structures but suffers from a heavy computational burden. One-shot NAS significantly alleviates the burden through weight sharing and improves computational efficiency. Zero-shot NAS further reduces the cost by predicting the performance of the network from its initial state, which conducts no training. Both methods aim to distinguish between "good" and "bad" architectures, i.e., ranking consistency of predicted and true performance. In this paper, we propose Ranking Distillation o…
End-to-end training with back propagation is the standard method for training deep neural networks. However, as networks become deeper and bigger, end-to-end training becomes more challenging: highly non-convex models gets stuck easily in local optima, gradients signals are prone to vanish or explode during back-propagation, training requires computational resources and time. In this work, we propose to break away from the end-to-end paradigm in the context of Knowledge Distillation. Instead of distilling a model end-to-end, we propose to split it into smaller sub-networks - also called neighb…
Infrared UAV Target Tracking with Dynamic Feature Refinement and Global Contextual Attention Knowledge Distillation
School of Computer Science and Technology, Xidian University（西安电子科技大学计算机科学与技术学院）；Xi’an Modern Control Technology Research Institute（西安现代控制技术研究所）；School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学人工智能与自动化学院）
University of Chinese Academy of Sciences（中国科学院大学）；Microsoft Research（微软研究院）；City University of Hong Kong（香港城市大学）；Stony Brook University（石溪大学）
Differentiable ARchiTecture Search, i.e., DARTS, has drawn great attention in neural architecture search. It tries to find the optimal architecture in a shallow search network and then measures its performance in a deep evaluation network. The independent optimization of the search and evaluation networks, however, leaves room for potential improvement by allowing interaction between the two networks. To address the problematic optimization issue, we propose new joint optimization objectives and a novel Cyclic Differentiable ARchiTecture Search framework, dubbed CDARTS. Considering the structu…
University of Toronto（多伦多大学）；University of New Brunswick（新不伦瑞克大学）
Neural Architecture Search (NAS) has been pivotal in finding optimal network configurations for Convolution Neural Networks (CNNs). While many methods explore NAS from a global search-space perspective, the employed optimization schemes typically require heavy computational resources. This work introduces a method that is efficient in computationally constrained environments by examining the micro-search space of channel size. In tackling channel-size optimization, we design an automated algorithm to extract the dependencies within different connected layers of the network. In addition, we int…
Several solutions for lightweight TTS have shown promising results. Still, they either rely on a hand-crafted design that reaches non-optimum size or use a neural architecture search but often suffer training costs. We present Nix-TTS, a lightweight TTS achieved via knowledge distillation to a high-quality yet large-sized, non-autoregressive, and end-to-end (vocoder-free) TTS teacher model. Specifically, we offer module-wise distillation, enabling flexible and independent distillation to the encoder and decoder module. The resulting Nix-TTS inherited the advantageous properties of being non-au…
INK: Inheritable Natural Backdoor Attack Against Model Distillation
Institute of Computer Application, China Academy of Engineering Physics（中国工程物理研究院计算机应用研究所）；Faculty of Electronic and Information Engineering, Xi’an Jiaotong University（西安交通大学电子与信息工程学院）；School of Cyberspace Security, Beijing University of Posts and Telecommunications（北京邮电大学网络空间安全学院）
针对可信数据集蒸馏可阻断传统投毒后门的问题，本文提出 INK，以自然图像方差为触发器，在无训练数据访问权时实现可从教师迁移到学生的干净图像、干净标签攻击，蒸馏后 ASR 仍超 98%。
Modern search systems use several large ranker models with transformer architectures. These models require large computational resources and are not suitable for usage on devices with limited computational resources. Knowledge distillation is a popular compression technique that can reduce the resource needs of such models, where a large teacher model transfers knowledge to a small student model. To drastically reduce memory requirements and energy consumption, we propose two extensions for a popular sentence-transformer distillation procedure: generation of an optimal size vocabulary and dime…
Qualcomm AI Research（高通人工智能研究院）；University of Amsterdam（阿姆斯特丹大学）
This paper aims to accelerate video stream processing, such as object detection and semantic segmentation, by leveraging the temporal redundancies that exist between video frames. Instead of propagating and warping features using motion alignment, such as optical flow, we propose a novel knowledge distillation schema coined as Delta Distillation. In our proposal, the student learns the variations in the teacher's intermediate features over time. We demonstrate that these temporal variations can be effectively distilled due to the temporal redundancies within video frames. During inference, bot…
University of Udine（乌迪内大学）；Ghent University - imec（根特大学 - imec）
Person re-identification (re-id) aims to retrieve images of same identities across different camera views. Resolution mismatch occurs due to varying distances between person of interest and cameras, this significantly degrades the performance of re-id in real world scenarios. Most of the existing approaches resolve the re-id task as low resolution problem in which a low resolution query image is searched in a high resolution images gallery. Several approaches apply image super resolution techniques to produce high resolution images but ignore the multiple resolutions of gallery images which is…
