Adversarial training methods typically align distributions by solving two-player games. However, in most current formulations, even if the generator aligns perfectly with data, a sub-optimal discriminator can still drive the two apart. Absent additional regularization, the instability can manifest itself as a never-ending game. In this paper, we introduce a family of objectives by leveraging pairwise discriminators, and show that only the generator needs to converge. The alignment, if achieved, would be preserved with any discriminator. We provide sufficient conditions for local convergence; characterize the capacity balance that should guide the discriminator and generator choices; and construct examples of minimally sufficient discriminators. Empirically, we illustrate the theory and the effectiveness of our approach on synthetic examples. Moreover, we show that practical methods derived from our approach can better generate higher-resolution images.
We study two important concepts in adversarial deep learning---adversarial training and generative adversarial network (GAN). Adversarial training is the technique used to improve the robustness of discriminator by combining adversarial attacker and discriminator in the training phase. GAN is commonly used for image generation by jointly optimizing discriminator and generator. We show these two concepts are indeed closely related and can be used to strengthen each other---adding a generator to the adversarial training procedure can improve the robustness of discriminators, and adding an advers…
UC San Diego（加州大学圣地亚哥分校）；SPAWAR（SPAWAR（太空与海战系统司令部））
We present a neural network architecture based upon the Autoencoder (AE) and Generative Adversarial Network (GAN) that promotes a convex latent distribution by training adversarially on latent space interpolations. By using an AE as both the generator and discriminator of a GAN, we pass a pixel-wise error function across the discriminator, yielding an AE which produces non-blurry samples that match both high- and low-level features of the original images. Interpolations between images in this space remain within the latent-space distribution of real images as trained by the discriminator, and …
Recent work has shown that state-of-the-art models are highly vulnerable to adversarial perturbations of the input. We propose cowboy, an approach to detecting and defending against adversarial attacks by using both the discriminator and generator of a GAN trained on the same dataset. We show that the discriminator consistently scores the adversarial samples lower than the real samples across multiple attacks and datasets. We provide empirical evidence that adversarial samples lie outside of the data manifold learned by the GAN. Based on this, we propose a cleaning method which uses both the d…
Québec AI institute (Mila)（魁北克人工智能研究所（Mila））；McGill University（麦吉尔大学）；Polytechnique Montréal（蒙特利尔理工学院）；Facebook AI Research（Facebook人工智能研究院）；Element AI；Ubisoft Montreal’s La Forge R&D laboratory（育碧蒙特利尔La Forge研发实验室）
Adversarial Imitation Learning alternates between learning a discriminator -- which tells apart expert's demonstrations from generated ones -- and a generator's policy to produce trajectories that can fool this discriminator. This alternated optimization is known to be delicate in practice since it compounds unstable adversarial training with brittle and sample-inefficient reinforcement learning. We propose to remove the burden of the policy optimization steps by leveraging a novel discriminator formulation. Specifically, our discriminator is explicitly conditioned on two policies: the one fro…
In recent years, Generative Adversarial Networks (GANs) have drawn a lot of attentions for learning the underlying distribution of data in various applications. Despite their wide applicability, training GANs is notoriously difficult. This difficulty is due to the min-max nature of the resulting optimization problem and the lack of proper tools of solving general (non-convex, non-concave) min-max optimization problems. In this paper, we try to alleviate this problem by proposing a new generative network that relies on the use of random discriminators instead of adversarial design. This design …
It is difficult to be able to imitate well in unknown states from a small amount of expert data and sampling data. Supervised learning methods such as Behavioral Cloning do not require sampling data, but usually suffer from distribution shift. The methods based on reinforcement learning, such as inverse reinforcement learning and generative adversarial imitation learning (GAIL), can learn from only a few expert data. However, they often need to interact with the environment. Soft Q imitation learning addressed the problems, and it was shown that it could learn efficiently by combining Behavior…
Consider the problem of simultaneous estimation of location and variance matrix under Huber's contaminated Gaussian model. First, we study minimum $f$-divergence estimation at the population level, corresponding to a generative adversarial method with a nonparametric discriminator and establish conditions on $f$-divergences which lead to robust estimation, similarly to robustness of minimum distance estimation. More importantly, we develop tractable adversarial algorithms with simple spline discriminators, which can be implemented via nested optimization such that the discriminator parameters …
This paper considers learning robot locomotion and manipulation tasks from expert demonstrations. Generative adversarial imitation learning (GAIL) trains a discriminator that distinguishes expert from agent transitions, and in turn use a reward defined by the discriminator output to optimize a policy generator for the agent. This generative adversarial training approach is very powerful but depends on a delicate balance between the discriminator and the generator training. In high-dimensional problems, the discriminator training may easily overfit or exploit associations with task-irrelevant f…
We introduce a novel method for emotion conversion in speech that does not require parallel training data. Our approach loosely relies on a cycle-GAN schema to minimize the reconstruction error from converting back and forth between emotion pairs. However, unlike the conventional cycle-GAN, our discriminator classifies whether a pair of input real and generated samples corresponds to the desired emotion conversion (e.g., A to B) or to its inverse (B to A). We will show that this setup, which we refer to as a variational cycle-GAN (VC-GAN), is equivalent to minimizing the empirical KL divergenc…
Institute of Automation（自动化研究所）；University of Chinese Academy of Sciences (UCAS)（中国科学院大学）；Alibaba Group（阿里巴巴集团）；Twitter（推特）
A fundamental challenge for machine learning models is how to generalize learned models for out-of-distribution (OOD) data. Among various approaches, exploiting invariant features by Domain Adversarial Training (DAT) received widespread attention. Despite its success, we observe training instability from DAT, mostly due to over-confident domain discriminator and environment label noise. To address this issue, we proposed Environment Label Smoothing (ELS), which encourages the discriminator to output soft probability, which thus reduces the confidence of the discriminator and alleviates the imp…
École de Technologie Supérieure (ÉTS), Université du Québec（魁北克大学高等技术学院）
This paper introduces a defense approach against end-to-end adversarial attacks developed for cutting-edge speech-to-text systems. The proposed defense algorithm has four major steps. First, we represent speech signals with 2D spectrograms using the short-time Fourier transform. Second, we iteratively find a safe vector using a spectrogram subspace projection operation. This operation minimizes the chordal distance adjustment between spectrograms with an additional regularization term. Third, we synthesize a spectrogram with such a safe vector using a novel GAN architecture trained with Sobole…
ST Electronics - SUTD Cyber Security Laboratory（新加坡科技电子-新加坡科技设计大学网络安全实验室）；Singapore University of Technology and Design（新加坡科技设计大学）
We propose to improve unconditional Generative Adversarial Networks (GAN) by training the self-supervised learning with the adversarial process. In particular, we apply self-supervised learning via the geometric transformation on input images and assign the pseudo-labels to these transformed images. (i) In addition to the GAN task, which distinguishes data (real) versus generated (fake) samples, we train the discriminator to predict the correct pseudo-labels of real transformed samples (classification task). Importantly, we find out that simultaneously training the discriminator to classify th…
Illinois Institute of Technology（伊利诺伊理工学院）；Washington University in St. Louis（圣路易斯华盛顿大学）
We introduce a generative adversarial network (GAN) model to simulate the 3-dimensional Lagrangian motion of particles trapped in the recirculation zone of a buoyancy-opposed flame. The GAN model comprises a stochastic recurrent neural network, serving as a generator, and a convoluted neural network, serving as a discriminator. Adversarial training was performed to the point where the best-trained discriminator failed to distinguish the ground truth from the trajectory produced by the best-trained generator. The model performance was then benchmarked against a statistical analysis performed on…
Institute for Artificial Intelligence（人工智能研究院）；State Key Lab of Intelligent Technology and Systems（智能技术与系统国家重点实验室）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；Tsinghua University（清华大学）
Most of the existing generative adversarial networks (GAN) for text generation suffer from the instability of reinforcement learning training algorithms such as policy gradient, leading to unstable performance. To tackle this problem, we propose a novel framework called Adversarial Reward Augmented Maximum Likelihood (ARAML). During adversarial training, the discriminator assigns rewards to samples which are acquired from a stationary distribution near the data rather than the generator's distribution. The generator is optimized with maximum likelihood estimation augmented by the discriminator…
We propose a novel training algorithm for a multi-speaker neural text-to-speech (TTS) model based on multi-task adversarial training. A conventional generative adversarial network (GAN)-based training algorithm significantly improves the quality of synthetic speech by reducing the statistical difference between natural and synthetic speech. However, the algorithm does not guarantee the generalization performance of the trained TTS model in synthesizing voices of unseen speakers who are not included in the training data. Our algorithm alternatively trains two deep neural networks: multi-task di…
We investigate under and overfitting in Generative Adversarial Networks (GANs), using discriminators unseen by the generator to measure generalization. We find that the model capacity of the discriminator has a significant effect on the generator's model quality, and that the generator's poor performance coincides with the discriminator underfitting. Contrary to our expectations, we find that generators with large model capacities relative to the discriminator do not show evidence of overfitting on CIFAR10, CIFAR100, and CelebA.
Adversarial learning methods have been proposed for a wide range of applications, but the training of adversarial models can be notoriously unstable. Effectively balancing the performance of the generator and discriminator is critical, since a discriminator that achieves very high accuracy will produce relatively uninformative gradients. In this work, we propose a simple and general technique to constrain information flow in the discriminator by means of an information bottleneck. By enforcing a constraint on the mutual information between the observations and the discriminator's internal repr…
Salesforce（赛富时）；Amazon（亚马逊）；Florida State University（佛罗里达州立大学）
The study of multidimensional discriminator (critic) output for Generative Adversarial Networks has been underexplored in the literature. In this paper, we generalize the Wasserstein GAN framework to take advantage of multidimensional critic output and explore its properties. We also introduce a square-root velocity transformation (SRVT) block which favors training in the multidimensional setting. Proofs of properties are based on our proposed maximal p-centrality discrepancy, which is bounded above by p-Wasserstein distance and fits the Wasserstein GAN framework with multidimensional critic o…
We propose to incorporate adversarial dropout in generative multi-adversarial networks, by omitting or dropping out, the feedback of each discriminator in the framework with some probability at the end of each batch. Our approach forces the single generator not to constrain its output to satisfy a single discriminator, but, instead, to satisfy a dynamic ensemble of discriminators. We show that this leads to a more generalized generator, promoting variety in the generated samples and avoiding the common mode collapse problem commonly experienced with generative adversarial networks (GANs). We f…
Unsupervised domain adaptation aims at transferring knowledge from the labeled source domain to the unlabeled target domain. Previous adversarial domain adaptation methods mostly adopt the discriminator with binary or $K$-dimensional output to perform marginal or conditional alignment independently. Recent experiments have shown that when the discriminator is provided with domain information in both domains and label information in the source domain, it is able to preserve the complex multimodal information and high semantic information in both domains. Following this idea, we adopt a discrimi…
INRS-EMT, Université du Québec（魁北克大学INRS-EMT分校）；Desautels Faculty of Management, McGill University（麦吉尔大学德索泰尔管理学院）；Quebec Artificial Intelligence Institute, Université de Montréal（蒙特利尔大学魁北克人工智能研究所）
Recent literature has demonstrated promising results for training Generative Adversarial Networks by employing a set of discriminators, in contrast to the traditional game involving one generator against a single adversary. Such methods perform single-objective optimization on some simple consolidation of the losses, e.g. an arithmetic average. In this work, we revisit the multiple-discriminator setting by framing the simultaneous minimization of losses provided by different models as a multi-objective optimization problem. Specifically, we evaluate the performance of multiple gradient descent…
Courant Institute of Mathematical Sciences, New York University（纽约大学柯朗数学科学研究所）；Google Brain（谷歌大脑）
We identify two issues with the family of algorithms based on the Adversarial Imitation Learning framework. The first problem is implicit bias present in the reward functions used in these algorithms. While these biases might work well for some environments, they can also lead to sub-optimal behavior in others. Secondly, even though these algorithms can learn from few expert demonstrations, they require a prohibitively large number of interactions with the environment in order to imitate the expert for many real-world applications. In order to address these issues, we propose a new algorithm c…
Anomaly detection is often considered a challenging field of machine learning due to the difficulty of obtaining anomalous samples for training and the need to obtain a sufficient amount of training data. In recent years, autoencoders have been shown to be effective anomaly detectors that train only on "normal" data. Generative adversarial networks (GANs) have been used to generate additional training samples for classifiers, thus making them more accurate and robust. However, in anomaly detection GANs are only used to reconstruct existing samples rather than to generate additional ones. This …
Recent work has shown that exploiting relations between labels improves the performance of multi-label classification. We propose a novel framework based on generative adversarial networks (GANs) to model label dependency. The discriminator learns to model label dependency by discriminating real and generated label sets. To fool the discriminator, the classifier, or generator, learns to generate label sets with dependencies close to real data. Extensive experiments and comparisons on two large-scale image classification benchmark datasets (MS-COCO and NUS-WIDE) show that the discriminator impr…
École Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）
The standard practice in Generative Adversarial Networks (GANs) discards the discriminator during sampling. However, this sampling method loses valuable information learned by the discriminator regarding the data distribution. In this work, we propose a collaborative sampling scheme between the generator and the discriminator for improved data generation. Guided by the discriminator, our approach refines the generated samples through gradient-based updates at a particular layer of the generator, shifting the generator distribution closer to the real data distribution. Additionally, we present …
Generative adversarial networks (GANs) have shown great success in applications such as image generation and inpainting. However, they typically require large datasets, which are often not available, especially in the context of prediction tasks such as image segmentation that require labels. Therefore, methods such as the CycleGAN use more easily available unlabelled data, but do not offer a way to leverage additional labelled data for improved performance. To address this shortcoming, we show how to factorise the joint data distribution into a set of lower-dimensional distributions along wit…
We introduce the Metropolis-Hastings generative adversarial network (MH-GAN), which combines aspects of Markov chain Monte Carlo and GANs. The MH-GAN draws samples from the distribution implicitly defined by a GAN's discriminator-generator pair, as opposed to standard GANs which draw samples from the distribution defined only by the generator. It uses the discriminator from GAN training to build a wrapper around the generator for improved sampling. With a perfect discriminator, this wrapped generator samples from the true distribution on the data exactly even when the generator is imperfect. W…
Generative Adversarial Networks (GANs) are one of the most popular tools for learning complex high dimensional distributions. However, generalization properties of GANs have not been well understood. In this paper, we analyze the generalization of GANs in practical settings. We show that discriminators trained on discrete datasets with the original GAN loss have poor generalization capability and do not approximate the theoretically optimal discriminator. We propose a zero-centered gradient penalty for improving the generalization of the discriminator by pushing it toward the optimal discrimin…
University of Louisville（路易斯维尔大学）；University of Social Science（社会科学大学）
The two key players in Generative Adversarial Networks (GANs), the discriminator and generator, are usually parameterized as deep neural networks (DNNs). On many generative tasks, GANs achieve state-of-the-art performance but are often unstable to train and sometimes miss modes. A typical failure mode is the collapse of the generator to a single parameter configuration where its outputs are identical. When this collapse occurs, the gradient of the discriminator may point in similar directions for many similar points. We hypothesize that some of these shortcomings are in part due to primitive a…
In this paper, we show that Generative Adversarial Networks (GANs) suffer from catastrophic forgetting even when they are trained to approximate a single target distribution. We show that GAN training is a continual learning problem in which the sequence of changing model distributions is the sequence of tasks to the discriminator. The level of mismatch between tasks in the sequence determines the level of forgetting. Catastrophic forgetting is interrelated to mode collapse and can make the training of GANs non-convergent. We investigate the landscape of the discriminator's output in different…
Generative Adversarial Networks (GANs) have achieved great success in generating realistic images. Most of these are conditional models, although acquisition of class labels is expensive and time-consuming in practice. To reduce the dependence on labeled data, we propose an un-conditional generative adversarial model, called K-Means-GAN (KM-GAN), which incorporates the idea of updating centers in K-Means into GANs. Specifically, we redesign the framework of GANs by applying K-Means on the features extracted from the discriminator. With obtained labels from K-Means, we propose new objective fun…
A recent technical breakthrough in the domain of machine learning is the discovery and the multiple applications of Generative Adversarial Networks (GANs). Those generative models are computationally demanding, as a GAN is composed of two deep neural networks, and because it trains on large datasets. A GAN is generally trained on a single server. In this paper, we address the problem of distributing GANs so that they are able to train over datasets that are spread on multiple workers. MD-GAN is exposed as the first solution for this problem: we propose a novel learning procedure for GANs so th…
A Triangle Generative Adversarial Network ($Δ$-GAN) is developed for semi-supervised cross-domain joint distribution matching, where the training data consists of samples from each domain, and supervision of domain correspondence is provided by only a few paired samples. $Δ$-GAN consists of four neural networks, two generators and two discriminators. The generators are designed to learn the two-way conditional distributions between the two domains, while the discriminators implicitly define a ternary discriminative function, which is trained to distinguish real data pairs and two kinds of fake…
Generative adversarial network (GAN) has become one of the most important neural network models for classical unsupervised machine learning. A variety of discriminator loss functions have been developed to train GAN's discriminators and they all have a common structure: a sum of real and fake losses that only depends on the actual and generated data respectively. One challenge associated with an equally weighted sum of two losses is that the training may benefit one loss but harm the other, which we show causes instability and mode collapse. In this paper, we introduce a new family of discrimi…
McGill University（麦吉尔大学）；Université du Québec（魁北克大学）；INRS-EMT（加拿大国家科学研究院能源、材料与电信研究所）；Mila – Quebec Artificial Intelligence Institute（米拉-魁北克人工智能研究所）；Facebook AI Research（脸书人工智能研究院）；Microsoft Research Montreal（微软蒙特利尔研究院）
Generative Adversarial Networks (GANs) can successfully approximate a probability distribution and produce realistic samples. However, open questions such as sufficient convergence conditions and mode collapse still persist. In this paper, we build on existing work in the area by proposing a novel framework for training the generator against an ensemble of discriminator networks, which can be seen as a one-student/multiple-teachers setting. We formalize this problem within the full-information adversarial bandit framework, where we evaluate the capability of an algorithm to select mixtures of …
Current approaches have made great progress on image-to-image translation tasks benefiting from the success of image synthesis methods especially generative adversarial networks (GANs). However, existing methods are limited to handling translation tasks between two species while keeping the content matching on the semantic level. A more challenging task would be the translation among more than two species. To explore this new area, we propose a simple yet effective structure of a multi-branch discriminator for enhancing an arbitrary generative adversarial architecture (GAN), named GAN-MBD. It …
We propose to tackle the mode collapse problem in generative adversarial networks (GANs) by using multiple discriminators and assigning a different portion of each minibatch, called microbatch, to each discriminator. We gradually change each discriminator's task from distinguishing between real and fake samples to discriminating samples coming from inside or outside its assigned microbatch by using a diversity parameter $α$. The generator is then forced to promote variety in each minibatch to make the microbatch discrimination harder to achieve by each discriminator. Thus, all models in our fr…
Filtering out unrealistic images from trained generative adversarial networks (GANs) has attracted considerable attention recently. Two density ratio based subsampling methods---Discriminator Rejection Sampling (DRS) and Metropolis-Hastings GAN (MH-GAN)---were recently proposed, and their effectiveness in improving GANs was demonstrated on multiple datasets. However, DRS and MH-GAN are based on discriminator based density ratio estimation (DRE) methods, so they may not work well if the discriminator in the trained GAN is far from optimal. Moreover, they do not apply to some GANs (e.g., MMD-GAN…
1 LINE Corp., Tokyo, Japan 2 NAVER Corp., Seongnam, Korea 3 Search Solutions Inc., Seongnam, Korea
This paper proposes voicing-aware conditional discriminators for Parallel WaveGAN-based waveform synthesis systems. In this framework, we adopt a projection-based conditioning method that can significantly improve the discriminator's performance. Furthermore, the conventional discriminator is separated into two waveform discriminators for modeling voiced and unvoiced speech. As each discriminator learns the distinctive characteristics of the harmonic and noise components, respectively, the adversarial training process becomes more efficient, allowing the generator to produce more realistic spe…
Mila；University of Montreal（蒙特利尔大学）；Facebook AI Research
Parametric adversarial divergences, which are a generalization of the losses used to train generative adversarial networks (GANs), have often been described as being approximations of their nonparametric counterparts, such as the Jensen-Shannon divergence, which can be derived under the so-called optimal discriminator assumption. In this position paper, we argue that despite being "non-optimal", parametric divergences have distinct properties from their nonparametric counterparts which can make them more suitable for learning high-dimensional distributions. A key property is that parametric di…
Computer Vision Center (CVC)（计算机视觉中心）；Universitat Autònoma de Barcelona（巴塞罗那自治大学）
Generative Adversarial Networks have long since revolutionized the world of computer vision and, tied to it, the world of art. Arduous efforts have gone into fully utilizing and stabilizing training so that outputs of the Generator network have the highest possible fidelity, but little has gone into using the Discriminator after training is complete. In this work, we propose to use the latter and show a way to use the features it has learned from the training dataset to both alter an image and generate one from scratch. We name this method Discriminator Dreaming, and the full code can be found…
In this paper, we present a conditional GAN with two generators and a common discriminator for multiview learning problems where observations have two views, but one of them may be missing for some of the training samples. This is for example the case for multilingual collections where documents are not available in all languages. Some studies tackled this problem by assuming the existence of view generation functions to approximately complete the missing views; for example Machine Translation to translate documents into the missing languages. These functions generally require an external reso…
Although GAN-based methods have received many achievements in the last few years, they have not been entirelysuccessful in generating discrete data. The most crucial challenge of these methods is the difficulty of passing the gradientfrom the discriminator to the generator when the generator outputs are discrete. Despite the fact that several attemptshave been made to alleviate this problem, none of the existing GAN-based methods have improved the performance oftext generation compared with the maximum likelihood approach in terms of both the quality and the diversity. In thispaper, we propose…
MILA（蒙特利尔学习算法研究所）；University of Montréal（蒙特利尔大学）；IVADO（魁北克人工智能研究所）；MSR（微软研究院）；University of Waterloo（滑铁卢大学）；New York University（纽约大学）；CIFAR（加拿大高级研究所）
Generative adversarial networks (GANs) are a learning framework that rely on training a discriminator to estimate a measure of difference between a target and generated distributions. GANs, as normally formulated, rely on the generated samples being completely differentiable w.r.t. the generative parameters, and thus do not work for discrete data. We introduce a method for training GANs with discrete data that uses the estimated difference measure from the discriminator to compute importance weights for generated samples, thus providing a policy gradient for training the generator. The importa…
In this paper, we extend the persona-based sequence-to-sequence (Seq2Seq) neural network conversation model to a multi-turn dialogue scenario by modifying the state-of-the-art hredGAN architecture to simultaneously capture utterance attributes such as speaker identity, dialogue topic, speaker sentiments and so on. The proposed system, phredGAN has a persona-based HRED generator (PHRED) and a conditional discriminator. We also explore two approaches to accomplish the conditional discriminator: (1) phredGAN_a, a system that passes the attribute representation as an additional input into a tradit…
NTT Communication Science Laboratories, NTT Corporation（NTT 公司 NTT 通信科学实验室）
In speech synthesis, a generative adversarial network (GAN), training a generator (speech synthesizer) and a discriminator in a min-max game, is widely used to improve speech quality. An ensemble of discriminators is commonly used in recent neural vocoders (e.g., HiFi-GAN) and end-to-end text-to-speech (TTS) systems (e.g., VITS) to scrutinize waveforms from multiple perspectives. Such discriminators allow synthesized speech to adequately approach real speech; however, they require an increase in the model size and computation time according to the increase in the number of discriminators. Alte…
In this paper, we propose Orthogonal Generative Adversarial Networks (O-GANs). We decompose the network of discriminator orthogonally and add an extra loss into the objective of common GANs, which can enforce discriminator become an effective encoder. The same extra loss can be embedded into any kind of GANs and there is almost no increase in computation. Furthermore, we discuss the principle of our method, which is relative to the fully-exploiting of the remaining degrees of freedom of discriminator. As we know, our solution is the simplest approach to train a generative adversarial network w…
