arXiv subjects

Yufei Wang

Publications and source records attributed to Yufei Wang.

At least 181 records · Page 10Linked to original sources

arXiv2020

f-IRL: Inverse Reinforcement Learning via State Marginal Matching

Imitation learning is well-suited for robotic tasks where it is difficult to directly program the behavior or specify a cost for optimal control. In this work, we propose a method for learning the reward function (and the corresponding policy) to match the expert state density. Our main result is the analytic gradient of any f-divergence between the agent and expert state distribution w.r.t. reward parameters. Based on the derived gradient, we present an algorithm, f-IRL, that recovers a stationary reward function from the expert density by gradient descent. We show that f-IRL can learn behaviors from a hand-designed target state density or implicitly through expert observations. Our method outperforms adversarial imitation learning methods in terms of sample efficiency and the required number of expert trajectories on IRL benchmarks. Moreover, we show that the recovered reward function can be used to quickly solve downstream tasks, and empirically demonstrate its utility on hard-to-explore tasks and for behavior transfer across changes in dynamics.

cs.LG↗

arXiv2020

ROLL: Visual Self-Supervised Reinforcement Learning with Object Reasoning

Current image-based reinforcement learning (RL) algorithms typically operate on the whole image without performing object-level reasoning. This leads to inefficient goal sampling and ineffective reward functions. In this paper, we improve upon previous visual self-supervised RL by incorporating object-level reasoning and occlusion reasoning. Specifically, we use unknown object segmentation to ignore distractors in the scene for better reward computation and goal generation; we further enable occlusion reasoning by employing a novel auxiliary loss and training scheme. We demonstrate that our proposed algorithm, ROLL (Reinforcement learning with Object Level Learning), learns dramatically faster and achieves better final performance compared with previous methods in several simulated visual control tasks. Project video and code are available at https://sites.google.com/andrew.cmu.edu/roll.

cs.LG↗

arXiv2020

Beyond Exponentially Discounted Sum: Automatic Learning of Return Function

In reinforcement learning, Return, which is the weighted accumulated future rewards, and Value, which is the expected return, serve as the objective that guides the learning of the policy. In classic RL, return is defined as the exponentially discounted sum of future rewards. One key insight is that there could be many feasible ways to define the form of the return function (and thus the value), from which the same optimal policy can be derived, yet these different forms might render dramatically different speeds of learning this policy. In this paper, we research how to modify the form of the return function to enhance the learning towards the optimal policy. We propose to use a general mathematical form for return function, and employ meta-learning to learn the optimal return function in an end-to-end manner. We test our methods on a specially designed maze environment and several Atari games, and our experimental results clearly indicate the advantages of automatically learning optimal return functions in reinforcement learning.

cs.LG↗

arXiv2020

Learning to Discretize: Solving 1D Scalar Conservation Laws via Deep Reinforcement Learning

Conservation laws are considered to be fundamental laws of nature. It has broad applications in many fields, including physics, chemistry, biology, geology, and engineering. Solving the differential equations associated with conservation laws is a major branch in computational mathematics. The recent success of machine learning, especially deep learning in areas such as computer vision and natural language processing, has attracted a lot of attention from the community of computational mathematics and inspired many intriguing works in combining machine learning with traditional methods. In this paper, we are the first to view numerical PDE solvers as an MDP and to use (deep) RL to learn new solvers. As proof of concept, we focus on 1-dimensional scalar conservation laws. We deploy the machinery of deep reinforcement learning to train a policy network that can decide on how the numerical solutions should be approximated in a sequential and spatial-temporal adaptive manner. We will show that the problem of solving conservation laws can be naturally viewed as a sequential decision-making process, and the numerical schemes learned in such a way can easily enforce long-term accuracy. Furthermore, the learned policy network is carefully designed to determine a good local discrete approximation based on the current state of the solution, which essentially makes the proposed method a meta-learning approach. In other words, the proposed method is capable of learning how to discretize for a given situation mimicking human experts. Finally, we will provide details on how the policy network is trained, how well it performs compared with some state-of-the-art numerical solvers such as WENO schemes, and supervised learning based approach L3D and PINN, and how well it generalizes.

cs.LG↗

arXiv2020

Light Can Hack Your Face! Black-box Backdoor Attack on Face Recognition Systems

Deep neural networks (DNN) have shown great success in many computer vision applications. However, they are also known to be susceptible to backdoor attacks. When conducting backdoor attacks, most of the existing approaches assume that the targeted DNN is always available, and an attacker can always inject a specific pattern to the training data to further fine-tune the DNN model. However, in practice, such attack may not be feasible as the DNN model is encrypted and only available to the secure enclave. In this paper, we propose a novel black-box backdoor attack technique on face recognition systems, which can be conducted without the knowledge of the targeted DNN model. To be specific, we propose a backdoor attack with a novel color stripe pattern trigger, which can be generated by modulating LED in a specialized waveform. We also use an evolutionary computing strategy to optimize the waveform for backdoor attack. Our backdoor attack can be conducted in a very mild condition: 1) the adversary cannot manipulate the input in an unnatural way (e.g., injecting adversarial noise); 2) the adversary cannot access the training database; 3) the adversary has no knowledge of the training model as well as the training set used by the victim party. We show that the backdoor trigger can be quite effective, where the attack success rate can be up to $88\%$ based on our simulation study and up to $40\%$ based on our physical-domain study by considering the task of face recognition and verification based on at most three-time attempts during authentication. Finally, we evaluate several state-of-the-art potential defenses towards backdoor attacks, and find that our attack can still be effective. We highlight that our study revealed a new physical backdoor attack, which calls for the attention of the security issue of the existing face recognition/verification techniques.

cs.CR↗

arXiv2020

Heterogeneous Domain Generalization via Domain Mixup

One of the main drawbacks of deep Convolutional Neural Networks (DCNN) is that they lack generalization capability. In this work, we focus on the problem of heterogeneous domain generalization which aims to improve the generalization capability across different tasks, which is, how to learn a DCNN model with multiple domain data such that the trained feature extractor can be generalized to supporting recognition of novel categories in a novel target domain. To solve this problem, we propose a novel heterogeneous domain generalization method by mixing up samples across multiple source domains with two different sampling strategies. Our experimental results based on the Visual Decathlon benchmark demonstrates the effectiveness of our proposed method. The code is released in \url{https://github.com/wyf0912/MIXALL}

cs.CV↗

arXiv2020

Meta-SAC: Auto-tune the Entropy Temperature of Soft Actor-Critic via Metagradient

Exploration-exploitation dilemma has long been a crucial issue in reinforcement learning. In this paper, we propose a new approach to automatically balance between these two. Our method is built upon the Soft Actor-Critic (SAC) algorithm, which uses an "entropy temperature" that balances the original task reward and the policy entropy, and hence controls the trade-off between exploitation and exploration. It is empirically shown that SAC is very sensitive to this hyperparameter, and the follow-up work (SAC-v2), which uses constrained optimization for automatic adjustment, has some limitations. The core of our method, namely Meta-SAC, is to use metagradient along with a novel meta objective to automatically tune the entropy temperature in SAC. We show that Meta-SAC achieves promising performances on several of the Mujoco benchmarking tasks, and outperforms SAC-v2 over 10% in one of the most challenging tasks, humanoid-v2.

cs.LG↗

arXiv2020

Neural Constituency Parsing of Speech Transcripts

This paper studies the performance of a neural self-attentive parser on transcribed speech. Speech presents parsing challenges that do not appear in written text, such as the lack of punctuation and the presence of speech disfluencies (including filled pauses, repetitions, corrections, etc.). Disfluencies are especially problematic for conventional syntactic parsers, which typically fail to find any EDITED disfluency nodes at all. This motivated the development of special disfluency detection systems, and special mechanisms added to parsers specifically to handle disfluencies. However, we show here that neural parsers can find EDITED disfluency nodes, and the best neural parsers find them with an accuracy surpassing that of specialized disfluency detection systems, thus making these specialized mechanisms unnecessary. This paper also investigates a modified loss function that puts more weight on EDITED nodes. It also describes tree-transformations that simplify the disfluency detection task by providing alternative encodings of disfluencies and syntactic information.

cs.CL↗

arXiv2020

Learning Enabled Dense Space-division Multiplexing through a Single Multimode Fibre

Space-division multiplexing is a promising technology in optical fibre communication to improve the transmission capacity of a single optical fibre. However, the number of channels that can be multiplexed is limited by the crosstalks between channels, and the multiplexing is only applied to few-mode or multi-core fibres. Here, we propose a high-spatial-density channel multiplexing framework employing deep learning for standard multimode fibres (MMF). We present a proof-of-concept experimental system, consisting of a single light source, a single digital-micromirror-device modulator, a single detection camera, and a deep convolutional neural network (CNN) to demonstrate up to 400-channel simultaneous data transmission with accuracy close to 100% over MMFs of different types, diameters and lengths. A novel scalable semi-supervised learning model is proposed to adapt the CNN to the time-varying MMF information channels in real-time, to overcome the environmental changes such as temperature variations and vibrations, and to reconstruct the input data from complex crosstalks among hundreds of channels. This deep-learning based approach is promising to maximize the use of the spatial dimension of MMFs, and to break the present number-of-channel limit in space-division multiplexing for future high-capacity MMF transmission data links.

physics.optics↗

arXiv2019

Period-doubling bifurcation of dissipative-soliton-resonance pulses in a passively mode-locked fiber laser

We report on the experimental observation of period-doubling bifurcation of dissipative-soliton-resonance (DSR) pulses in a fiber laser passively mode-locked by using the nonlinear optical loop mirror. Increasing the pump power of the fiber laser, we show that temporally a stable, uniform DSR pulse train could be transformed into a period-doubling state, exhibiting two sets of pulse parameters between the adjacent cavity roundtrip. It is found that DSR pulses in the period-doubling state could maintain the typical feature of DSR pulse: fixed pulse peak power and linear variation in pulse width with respect to the pump power change. The mechanism for achieving period-doubling of DSR pulses is discussed.

physics.optics↗

arXiv2019

A game method for improving the interpretability of convolution neural network

Real artificial intelligence always has been focused on by many machine learning researchers, especially in the area of deep learning. However deep neural network is hard to be understood and explained, and sometimes, even metaphysics. The reason is, we believe that: the network is essentially a perceptual model. Therefore, we believe that in order to complete complex intelligent activities from simple perception, it is necessary to con-struct another interpretable logical network to form accurate and reasonable responses and explanations to external things. Researchers like Bolei Zhou and Quanshi Zhang have found many explanatory rules for deep feature extraction aimed at the feature extraction stage of convolution neural network. However, although researchers like Marco Gori have also made great efforts to improve the interpretability of the fully connected layers of the network, the problem is also very difficult. This paper firstly analyzes its reason. Then a method of constructing logical network based on the fully connected layers and extracting logical relation between input and output of the layers is proposed. The game process between perceptual learning and logical abstract cognitive learning is implemented to improve the interpretable performance of deep learning process and deep learning model. The benefits of our approach are illustrated on benchmark data sets and in real-world experiments.

cs.LG↗

arXiv2019

How to improve the interpretability of kernel learning

In recent years, machine learning researchers have focused on methods to construct flexible and interpretable prediction models. However, an interpretability evaluation, a relationship between generalization performance and an interpretability of the model and a method for improving the interpretability have to be considered. In this paper, a quantitative index of the interpretability is proposed and its rationality is proved, and equilibrium problem between the interpretability and the generalization performance is analyzed. Probability upper bound of the sum of the two performances is analyzed. For traditional supervised kernel machine learning problem, a universal learning framework is put forward to solve the equilibrium problem between the two performances. The condition for global optimal solution based on the framework is deduced. The learning framework is applied to the least-squares support vector machine and is evaluated by some experiments.

cs.LG↗

arXiv2019

nocaps: novel object captioning at scale

Image captioning models have achieved impressive results on datasets containing limited visual concepts and large amounts of paired image-caption training data. However, if these models are to ever function in the wild, a much larger variety of visual concepts must be learned, ideally from less supervision. To encourage the development of image captioning models that can learn visual concepts from alternative data sources, such as object detection datasets, we present the first large-scale benchmark for this task. Dubbed 'nocaps', for novel object captioning at scale, our benchmark consists of 166,100 human-generated captions describing 15,100 images from the OpenImages validation and test sets. The associated training data consists of COCO image-caption pairs, plus OpenImages image-level labels and object bounding boxes. Since OpenImages contains many more classes than COCO, nearly 400 object classes seen in test images have no or very few associated training captions (hence, nocaps). We extend existing novel object captioning models to establish strong baselines for this benchmark and provide analysis to guide future work on this task.

cs.CV↗

arXiv2019

UniDual: A Unified Model for Image and Video Understanding

Although a video is effectively a sequence of images, visual perception systems typically model images and videos separately, thus failing to exploit the correlation and the synergy provided by these two media. While a few prior research efforts have explored the benefits of leveraging still-image datasets for video analysis, or vice-versa, most of these attempts have been limited to pretraining a model on one type of visual modality and then adapting it via finetuning on the other modality. In contrast, in this paper we introduce a framework that enables joint training of a unified model on mixed collections of image and video examples spanning different tasks. The key ingredient in our architecture design is a new network block, which we name UniDual. It consists of a shared 2D spatial convolution followed by two parallel point-wise convolutional layers, one devoted to images and the other one used for videos. For video input, the point-wise filtering implements a temporal convolution. For image input, it performs a pixel-wise nonlinear transformation. Repeated stacking of such blocks gives rise to a network where images and videos undergo partially distinct execution pathways, unified by spatial convolutions (capturing commonalities in visual appearance) but separated by point-wise operations (modeling patterns specific to each modality). Extensive experiments on Kinetics and ImageNet demonstrate that our UniDual model jointly trained on these datasets yields substantial accuracy gains for both tasks, compared to 1) training separate models, 2) traditional multi-task learning and 3) the conventional framework of pretraining-followed-by-finetuning. On Kinetics, the UniDual architecture applied to a state-of-the-art video backbone model (R(2+1)D-152) yields an additional video@1 accuracy gain of 1.5%.

cs.CV↗

arXiv2019

How to best use Syntax in Semantic Role Labelling

There are many different ways in which external information might be used in an NLP task. This paper investigates how external syntactic information can be used most effectively in the Semantic Role Labeling (SRL) task. We evaluate three different ways of encoding syntactic parses and three different ways of injecting them into a state-of-the-art neural ELMo-based SRL sequence labelling model. We show that using a constituency representation as input features improves performance the most, achieving a new state-of-the-art for non-ensemble SRL models on the in-domain CoNLL'05 and CoNLL'12 benchmarks.

cs.CL↗

arXiv2018

How far from automatically interpreting deep learning

In recent years, deep learning researchers have focused on how to find the interpretability behind deep learning models. However, today cognitive competence of human has not completely covered the deep learning model. In other words, there is a gap between the deep learning model and the cognitive mode. How to evaluate and shrink the cognitive gap is a very important issue. In this paper, the interpretability evaluation, the relationship between the generalization performance and the interpretability of the model and the method for improving the interpretability are concerned. A universal learning framework is put forward to solve the equilibrium problem between the two performances. The uniqueness of solution of the problem is proved and condition of unique solution is obtained. Probability upper bound of the sum of the two performances is analyzed.

cs.LG↗

arXiv2018

Deep Reinforcement Learning for Green Security Games with Real-Time Information

Green Security Games (GSGs) have been proposed and applied to optimize patrols conducted by law enforcement agencies in green security domains such as combating poaching, illegal logging and overfishing. However, real-time information such as footprints and agents' subsequent actions upon receiving the information, e.g., rangers following the footprints to chase the poacher, have been neglected in previous work. To fill the gap, we first propose a new game model GSG-I which augments GSGs with sequential movement and the vital element of real-time information. Second, we design a novel deep reinforcement learning-based algorithm, DeDOL, to compute a patrolling strategy that adapts to the real-time information against a best-responding attacker. DeDOL is built upon the double oracle framework and the policy-space response oracle, solving a restricted game and iteratively adding best response strategies to it through training deep Q-networks. Exploring the game structure, DeDOL uses domain-specific heuristic strategies as initial strategies and constructs several local modes for efficient and parallelized training. To our knowledge, this is the first attempt to use Deep Q-Learning for security games.

cs.MA↗

arXiv2018

Novel development of dissipative-soliton-resonance pulses with pump power in an all-normal-dispersion fiber laser

Evolution of dissipative-soliton-resonance (DSR) pulses in an all-normal-dispersion fiber laser with pump power under various operation conditions are experimentally studied. The fiber laser is mode-locked by using a nonlinear fiber loop mirror. Apart from the typical pulse broadening due to the peak power clamping effect, pulse breaking was observed. In addition, pulse narrowing with pump power increasing was observed. All the incompatible evolution of DSR pulses could be attributed to the multiple parameters change due to pump power increasing under specific operation conditions.

physics.optics↗