Search

Time series classification models have been garnering significant importance in the research community. However, not much research has been done on generating adversarial samples for these models. These adversarial samples can become a security concern. In this paper, we propose utilizing an adversarial transformation network (ATN) on a distilled model to attack various time series classification models. The proposed attack on the classification model utilizes a distilled model as a surrogate that mimics the behavior of the attacked classical time series classification models. Our proposed methodology is applied onto 1-nearest neighbor dynamic time warping (1-NN DTW) and a fully convolutional network (FCN), all of which are trained on 42 University of California Riverside (UCR) datasets. In this paper, we show both models were susceptible to attacks on all 42 datasets. When compared to Fast Gradient Sign Method, the proposed attack generates a larger faction of successful adversarial black-box attacks. A simple defense mechanism is successfully devised to reduce the fraction of successful adversarial samples. Finally, we recommend future researchers that develop time series classification models to incorporating adversarial data samples into their training data sets to improve resilience on adversarial samples. [ABSTRACT FROM AUTHOR]

The heterogeneity in recently published knowledge graph embedding models’ implementations, training, and evaluation has made fair and thorough comparisons difficult. To assess the reproducibility of previously published results, we re-implemented and evaluated 21 models in the PyKEEN software package. In this paper, we outline which results could be reproduced with their reported hyper-parameters, which could only be reproduced with alternate hyper-parameters, and which could not be reproduced at all, as well as provide insight as to why this might be the case. We then performed a large-scale benchmarking on four datasets with several thousands of experiments and 24,804 GPU hours of computation time. We present insights gained as to best practices, best configurations for each model, and where improvements could be made over previously published best configurations. Our results highlight that the combination of model architecture, training approach, loss function, and the explicit modeling of inverse relations is crucial for a model’s performance and is not only determined by its architecture. We provide evidence that several architectures can obtain results competitive to the state of the art when configured carefully. We have made all code, experimental configurations, results, and analyses available at https://github.com/pykeen/pykeen and https://github.com/pykeen/benchmarking. [ABSTRACT FROM AUTHOR]

Intrinsic image decomposition is the task of mapping image to albedo and shading. Classical approaches derive methods from spatial models. The modern literature stresses evaluation, by comparing predictions to human judgements (“lighter”, “same as”, “darker”). The best modern intrinsic image methods train a map from image to albedo using images rendered from computer graphics models and example human judgements. This approach yields practical methods, but obtaining rendered images can be inconvenient. Furthermore, the approach cannot explain how a one could learn to recover intrinsic images without geometric, surface and illumination models, as people and animals appear to do. This paper describes a method that learns intrinsic image decomposition without seeing human annotations, rendered data, or ground truth data. Instead, the method relies on paradigms – spatial models of albedo and of shading. Rather than finding the “best” albedo and shading for an image via optimization, our approach trains a neural network on synthetic images. The synthetic images are constructed by multiplying albedos and shading fields sampled from our models. The network is subject to a novel smoothing procedure that ensures good behavior at short scales on real images. An averaging procedure ensures that reported albedo and shading are largely equivariant – different crops and scalings of an image will report the same albedo and shading at shared points. This averaging procedure controls long scale error. The standard evaluation for an intrinsic image method is a WHDR score. Our method achieves WHDR scores competitive with those of strong recent methods allowed to see training WHDR annotations, rendered data, and ground truth data. Our method produces albedo and shading maps with attractive qualitative properties – for example, albedo fields do not suppress wood grain and represent narrow grooves in surfaces well. Because our method is unsupervised, we can compute estimates of the test/train variance of WHDR scores; these are quite large, and suggest is unsafe to rely small differences in reported WHDR. [ABSTRACT FROM AUTHOR]

Pretrained deep models hold their learnt knowledge in the form of model parameters. These parameters act as “memory” for the trained models and help them generalize well on unseen data. However, in absence of training data, the utility of a trained model is merely limited to either inference or better initialization towards a target task. In this paper, we go further and extract synthetic data by leveraging the learnt model parameters. We dub them Data Impressions, which act as proxy to the training data and can be used to realize a variety of tasks. These are useful in scenarios where only the pretrained models are available and the training data is not shared (e.g., due to privacy or sensitivity concerns). We show the applicability of data impressions in solving several computer vision tasks such as unsupervised domain adaptation, continual learning as well as knowledge distillation. We also study the adversarial robustness of lightweight models trained via knowledge distillation using these data impressions. Further, we demonstrate the efficacy of data impressions in generating data-free Universal Adversarial Perturbations (UAPs) with better fooling rates. Extensive experiments performed on benchmark datasets demonstrate competitive performance achieved using data impressions in absence of original training data. [ABSTRACT FROM AUTHOR]

Traditional Deep Neural Network (DNN) security is mostly related to the well-known adversarial input example attack. Recently, another dimension of adversarial attack, namely, attack on DNN weight parameters, has been shown to be very powerful. As a representative one, the Bit-Flip-based adversarial weight Attack (BFA) injects an extremely small amount of faults into weight parameters to hijack the executing DNN function. Prior works of BFA focus on un-targeted attack that can hack all inputs into a random output class by flipping a very small number of weight bits stored in computer memory. This paper proposes the first work of targeted BFA based (T-BFA) adversarial weight attack on DNNs, which can intentionally mislead selected inputs to a target output class. The objective is achieved by identifying the weight bits that are highly associated with classification of a targeted output through a class-dependent vulnerable weight bit searching algorithm. Our proposed T-BFA performance is successfully demonstrated on multiple DNN architectures for image classification tasks. For example, by merely flipping 27 out of 88 million weight bits of ResNet-18, our T-BFA can misclassify all the images from ’Hen’ class into ’Goose’ class (i.e., 100% attack success rate) in ImageNet dataset, while maintaining 59.35% validation accuracy. Moreover, we successfully demonstrate our T-BFA attack in a real computer prototype system running DNN computation, with Ivy Bridge-based Intel i7 CPU and 8GB DDR3 memory. [ABSTRACT FROM AUTHOR]

As deep learning models are usually massive and complex, distributed learning is essential for increasing training efficiency. Moreover, in many real-world application scenarios like healthcare, distributed learning can also keep the data local and protect privacy. Recently, the asynchronous decentralized parallel stochastic gradient descent (ADPSGD) algorithm has been proposed and demonstrated to be an efficient and practical strategy where there is no central server, so that each computing node only communicates with its neighbors. Although no raw data will be transmitted across different local nodes, there is still a risk of information leak during the communication process for malicious participants to make attacks. In this paper, we present a differentially private version of asynchronous decentralized parallel SGD framework, or A(DP) $^2$ 2 SGD for short, which maintains communication efficiency of ADPSGD and prevents the inference from malicious participants. Specifically, Rényi differential privacy is used to provide tighter privacy analysis for our composite Gaussian mechanisms while the convergence rate is consistent with the non-private version. Theoretical analysis shows A(DP) $^2$ 2 SGD also converges at the optimal $\mathcal {O}(1/\sqrt{T})$ O (1 / T) rate as SGD. Empirically, A(DP) $^2$ 2 SGD achieves comparable model accuracy as the differentially private version of Synchronous SGD (SSGD) but runs much faster than SSGD in heterogeneous computing environments. [ABSTRACT FROM AUTHOR]

Click-through rate (CTR) prediction, whose goal is to predict the probability of the user to click on an item, has become increasingly significant in the recommender systems. Recently, some deep learning models with the ability to automatically extract the user interest from his/her behaviors have achieved great success. In these work, the attention mechanism is used to select the user interested items in historical behaviors, improving the performance of the CTR predictor. Normally, these attentive modules can be jointly trained with the base predictor by using gradient descents. In this paper, we regard user interest modeling as a feature selection problem, which we call user interest selection. For such a problem, we propose a novel approach under the framework of the wrapper method, which is named Meta-Wrapper. More specifically, we use a differentiable module as our wrapping operator and then recast its learning problem as a continuous bilevel optimization. Moreover, we use a meta-learning algorithm to solve the optimization and theoretically prove its convergence. Meanwhile, we also provide theoretical analysis to show that our proposed method 1) efficiencies the wrapper-based feature selection, and 2) achieves better resistance to overfitting. Finally, extensive experiments on three public datasets manifest the superiority of our method in boosting the performance of CTR prediction. [ABSTRACT FROM AUTHOR]

With substantial amount of time, resources and human (team) efforts invested to explore and develop successful deep neural networks (DNN), there emerges an urgent need to protect these inventions from being illegally copied, redistributed, or abused without respecting the intellectual properties of legitimate owners. Following recent progresses along this line, we investigate a number of watermark-based DNN ownership verification methods in the face of ambiguity attacks, which aim to cast doubts on the ownership verification by forging counterfeit watermarks. It is shown that ambiguity attacks pose serious threats to existing DNN watermarking methods. As remedies to the above-mentioned loophole, this paper proposes novel passport-based DNN ownership verification schemes which are both robust to network modifications and resilient to ambiguity attacks. The gist of embedding digital passports is to design and train DNN models in a way such that, the DNN inference performance of an original task will be significantly deteriorated due to forged passports. In other words, genuine passports are not only verified by looking for the predefined signatures, but also reasserted by the unyielding DNN model inference performances. Extensive experimental results justify the effectiveness of the proposed passport-based DNN ownership verification schemes. Code is available at https://github.com/kamwoh/DeepIPR. [ABSTRACT FROM AUTHOR]

Dynamic routing networks, aimed at finding the best routing paths in the networks, have achieved significant improvements to neural networks in terms of accuracy and efficiency. In this paper, we see dynamic routing networks in a fresh light, formulating a routing method as a mapping from a sample space to a routing space. From the perspective of space mapping, prevalent methods of dynamic routing did not take into account how routing paths would be distributed in the routing space. Thus, we propose a novel method, termed CoDiNet, to model the relationship between a sample space and a routing space by regularizing the distribution of routing paths with the properties of consistency and diversity. In principle, the routing paths for the self-supervised similar samples should be closely distributed in the routing space.Moreover, we design a customizable dynamic routing module, which can strike a balance between accuracy and efficiency. When deployed upon ResNet models, our method achieves higher performance and effectively reduces average computational cost on four widely used datasets. [ABSTRACT FROM AUTHOR]

We study network pruning which aims to remove redundant channels/kernels and hence speed up the inference of deep networks. Existing pruning methods either train from scratch with sparsity constraints or minimize the reconstruction error between the feature maps of the pre-trained models and the compressed ones. Both strategies suffer from some limitations: the former kind is computationally expensive and difficult to converge, while the latter kind optimizes the reconstruction error but ignores the discriminative power of channels. In this paper, we propose a simple-yet-effective method called discrimination-aware channel pruning (DCP) to choose the channels that actually contribute to the discriminative power. To this end, we first introduce additional discrimination-aware losses into the network to increase the discriminative power of the intermediate layers. Next, we select the most discriminative channels for each layer by considering the discrimination-aware loss and the reconstruction error, simultaneously. We then formulate channel pruning as a sparsity-inducing optimization problem with a convex objective and propose a greedy algorithm to solve the resultant problem. Note that a channel (3D tensor) often consists of a set of kernels (each with a 2D matrix). Besides the redundancy in channels, some kernels in a channel may also be redundant and fail to contribute to the discriminative power of the network, resulting in kernel level redundancy. To solve this issue, we propose a discrimination-aware kernel pruning (DKP) method to further compress deep networks by removing redundant kernels. To avoid manually determining the pruning rate for each layer, we propose two adaptive stopping conditions to automatically determine the number of selected channels/kernels. The proposed adaptive stopping conditions tend to yield more efficient models with better performance in practice. Extensive experiments on both image classification and face recognition demonstrate the effectiveness of our methods. For example, on ILSVRC-12, the resultant ResNet-50 model with 30 percent reduction of channels even outperforms the baseline model by 0.36 percent in terms of Top-1 accuracy. We also deploy the pruned models on a smartphone (equipped with a Qualcomm Snapdragon 845 processor). The pruned MobileNetV1 and MobileNetV2 achieve 1.93× and 1.42× inference acceleration on the mobile device, respectively, with negligible performance degradation. The source code and the pre-trained models are available at https://github.com/SCUT-AILab/DCP. [ABSTRACT FROM AUTHOR]

Remarkable achievements have been obtained by deep neural networks in the last several years. However, the breakthrough in neural networks accuracy is always accompanied by explosive growth of computation and parameters, which leads to a severe limitation of model deployment. In this paper, we propose a novel knowledge distillation technique named self-distillation to address this problem. Self-distillation attaches several attention modules and shallow classifiers at different depths of neural networks and distills knowledge from the deepest classifier to the shallower classifiers. Different from the conventional knowledge distillation methods where the knowledge of the teacher model is transferred to another student model, self-distillation can be considered as knowledge transfer in the same model - from the deeper layers to the shallow layers. Moreover, the additional classifiers in self-distillation allow the neural network to work in a dynamic manner, which leads to a much higher acceleration. Experiments demonstrate that self-distillation has consistent and significant effectiveness on various neural networks and datasets. On average, 3.49 and 2.32 percent accuracy boost are observed on CIFAR100 and ImageNet. Besides, experiments show that self-distillation can be combined with other model compression methods, including knowledge distillation, pruning and lightweight model design. [ABSTRACT FROM AUTHOR]

Text encoding is one of the most important steps in Natural Language Processing (NLP). It has been done well by the self-attention mechanism in the current state-of-the-art Transformer encoder, which has brought about significant improvements in the performance of many NLP tasks. Though the Transformer encoder may effectively capture general information in its resulting representations, the backbone information, meaning the gist of the input text, is not specifically focused on. In this paper, we propose explicit and implicit text compression approaches to enhance the Transformer encoding and evaluate models using this approach on several typical downstream tasks that rely on the encoding heavily. Our explicit text compression approaches use dedicated models to compress text, while our implicit text compression approach simply adds an additional module to the main model to handle text compression. We propose three ways of integration, namely backbone source-side fusion, target-side fusion, and both-side fusion, to integrate the backbone information into Transformer-based models for various downstream tasks. Our evaluation on benchmark datasets shows that the proposed explicit and implicit text compression approaches improve results in comparison to strong baselines. We therefore conclude, when comparing the encodings to the baseline models, text compression helps the encoders to learn better language representations. [ABSTRACT FROM AUTHOR]

Kernel methods have achieved tremendous success in the past two decades. In the current big data era, data collection has grown tremendously. However, existing kernel methods are not scalable enough both at the training and predicting steps. To address this challenge, in this paper, we first introduce a general sparse kernel learning formulation based on the random feature approximation, where the loss functions are possibly non-convex. In order to reduce the scale of random features required in experiment, we also use that formulation based on the orthogonal random feature approximation. Then we propose a new asynchronous parallel doubly stochastic algorithm for large scale sparse kernel learning (AsyDSSKL). To the best our knowledge, AsyDSSKL is the first algorithm with the techniques of asynchronous parallel computation and doubly stochastic optimization. We also provide a comprehensive convergence guarantee to AsyDSSKL. Importantly, the experimental results on various large-scale real-world datasets show that, our AsyDSSKL method has the significant superiority on the computational efficiency at the training and predicting steps over the existing kernel methods. [ABSTRACT FROM AUTHOR]

Deep neural models, in recent years, have been successful in almost every field, even solving the most complex problem statements. However, these models are huge in size with millions (and even billions) of parameters, demanding heavy computation power and failing to be deployed on edge devices. Besides, the performance boost is highly dependent on redundant labeled data. To achieve faster speeds and to handle the problems caused by the lack of labeled data, knowledge distillation (KD) has been proposed to transfer information learned from one model to another. KD is often characterized by the so-called ‘Student-Teacher’ (S-T) learning framework and has been broadly applied in model compression and knowledge transfer. This paper is about KD and S-T learning, which are being actively studied in recent years. First, we aim to provide explanations of what KD is and how/why it works. Then, we provide a comprehensive survey on the recent progress of KD methods together with S-T frameworks typically used for vision tasks. In general, we investigate some fundamental questions that have been driving this research area and thoroughly generalize the research progress and technical details. Additionally, we systematically analyze the research status of KD in vision applications. Finally, we discuss the potentials and open challenges of existing methods and prospect the future directions of KD and S-T learning. [ABSTRACT FROM AUTHOR]

The ability to predict, anticipate and reason about future outcomes is a key component of intelligent decision-making systems. In light of the success of deep learning in computer vision, deep-learning-based video prediction emerged as a promising research direction. Defined as a self-supervised learning task, video prediction represents a suitable framework for representation learning, as it demonstrated potential capabilities for extracting meaningful representations of the underlying patterns in natural videos. Motivated by the increasing interest in this task, we provide a review on the deep learning methods for prediction in video sequences. We first define the video prediction fundamentals, as well as mandatory background concepts and the most used datasets. Next, we carefully analyze existing video prediction models organized according to a proposed taxonomy, highlighting their contributions and their significance in the field. The summary of the datasets and methods is accompanied with experimental results that facilitate the assessment of the state of the art on a quantitative basis. The paper is summarized by drawing some general conclusions, identifying open research challenges and by pointing out future research directions. [ABSTRACT FROM AUTHOR]

The explosive growth in video streaming requires video understanding at high accuracy and low computation cost. Conventional 2D CNNs are computationally cheap but cannot capture temporal relationships; 3D CNN based methods can achieve good performance but are computationally intensive. In this paper, we propose a generic and effective Temporal Shift Module (TSM) that enjoys both high efficiency and high performance. The key idea of TSM is to shift part of the channels along the temporal dimension, thus facilitate information exchanged among neighboring frames. It can be inserted into 2D CNNs to achieve temporal modeling at zero computation and zero parameters. TSM offers several unique advantages. First, TSM has high performance; it ranks the first on the Something-Something leaderboard upon submission. Second, TSM has high efficiency; it achieves a high frame rate of 74fps and 29fps for online video recognition on Jetson Nano and Galaxy Note8. Third, TSM has higher scalability compared to 3D networks, enabling large-scale Kinetics training on 1,536 GPUs in 15 minutes. Lastly, TSM enables action concepts learning, which 2D networks cannot model; we visualize the category attention map and find that spatial-temporal action detector emerges during the training of classification tasks. The code is publicly available at https://github.com/mit-han-lab/temporal-shift-module. [ABSTRACT FROM AUTHOR]

Many real–world domains involve information naturally represented by graphs, where nodes denote basic patterns while edges stand for relationships among them. The graph neural network (GNN) is a machine learning model capable of directly managing graph–structured data. In the original framework, GNNs are inductively trained, adapting their parameters based on a supervised learning environment. However, GNNs can also take advantage of transductive learning, thanks to the natural way they make information flow and spread across the graph, using relationships among patterns. In this paper, we propose a mixed inductive–transductive GNN model, study its properties and introduce an experimental strategy that allows us to understand and distinguish the role of inductive and transductive learning. The preliminary experimental results show interesting properties for the mixed model, highlighting how the peculiarities of the problems and the data can impact on the two learning strategies. [ABSTRACT FROM AUTHOR]

Diversity is one of the most important properties in image captioning, as it reflects various expressions of important concepts presented in an image. However, the most popular metrics cannot well evaluate the diversity of multiple captions. In this paper, we first propose a metric to measure the diversity of a set of captions, which is derived from latent semantic analysis (LSA), and then kernelize LSA using CIDEr (R. Vedantam et al., 2015) similarity. Compared with mBLEU (R. Shetty et al., 2017), our proposed diversity metrics show a relatively strong correlation to human evaluation. We conduct extensive experiments, finding there is a large gap between the performance of the current state-of-the-art models and human annotations considering both diversity and accuracy; the models that aim to generate captions with higher CIDEr scores normally obtain lower diversity scores, which generally learn to describe images using common words. To bridge this “diversity” gap, we consider several methods for training caption models to generate diverse captions. First, we show that balancing the cross-entropy loss and CIDEr reward in reinforcement learning during training can effectively control the tradeoff between diversity and accuracy of the generated captions. Second, we develop approaches that directly optimize our diversity metric and CIDEr score using reinforcement learning. These proposed approaches using reinforcement learning (RL) can be unified into a self-critical (S. J. Rennie et al., 2017) framework with new RL baselines. Third, we combine accuracy and diversity into a single measure using an ensemble matrix, and then maximize the determinant of the ensemble matrix via reinforcement learning to boost diversity and accuracy, which outperforms its counterparts on the oracle test. Finally, inspired by determinantal point processes (DPP), we develop a DPP selection algorithm to select a subset of captions from a large number of candidate captions. The experimental results show that maximizing the determinant of the ensemble matrix outperforms other methods considerably improving diversity and accuracy. [ABSTRACT FROM AUTHOR]

Stochastic gradient descent (SGD) has become the method of choice for training highly complex and nonconvex models since it can not only recover good solutions to minimize training errors but also generalize well. Computational and statistical properties are separately studied to understand the behavior of SGD in the literature. However, there is a lacking study to jointly consider the computational and statistical properties in a nonconvex learning setting. In this paper, we develop novel learning rates of SGD for nonconvex learning by presenting high-probability bounds for both computational and statistical errors. We show that the complexity of SGD iterates grows in a controllable manner with respect to the iteration number, which sheds insights on how an implicit regularization can be achieved by tuning the number of passes to balance the computational and statistical errors. As a byproduct, we also slightly refine the existing studies on the uniform convergence of gradients by showing its connection to Rademacher chaos complexities. [ABSTRACT FROM AUTHOR]

This paper analyzes regularization terms proposed recently for improving the adversarial robustness of deep neural networks (DNNs), from a theoretical point of view. Specifically, we study possible connections between several effective methods, including input-gradient regularization, Jacobian regularization, curvature regularization, and a cross-Lipschitz functional. We investigate them on DNNs with general rectified linear activations, which constitute one of the most prevalent families of models for image classification and a host of other machine learning applications. We shed light on essential ingredients of these regularizations and re-interpret their functionality. Through the lens of our study, more principled and efficient regularizations can possibly be invented in the near future. [ABSTRACT FROM AUTHOR]