Graphical Minimax Game and Off-Policy Reinforcement Learning for Heterogeneous MASs with Spanning Tree Condition

In this paper, the optimal consensus control problem is investigated for heterogeneous linear multi-agent systems (MASs) with spanning tree condition based on game theory and reinforcement learning. First, the graphical minimax game algebraic Riccati equation (ARE) is derived by converting the consensus problem into a zero-sum game problem between each agent and its neighbors. The asymptotic stability and minimax validation of the closed-loop systems are proved theoretically. Then, a data-driven off-policy reinforcement learning algorithm is proposed to online learn the optimal control policy without the information of the system dynamics. A certain rank condition is established to guarantee the convergence of the proposed algorithm to the unique solution of the ARE. Finally, the effectiveness of the proposed method is demonstrated through a numerical simulation.

IGNITE: A Minimax Game Toward Learning Individual Treatment Effects from Networked Observational Data

Networked observational data presents new opportunities for learning individual causal effects, which plays an indispensable role in decision making. Such data poses the challenge of confounding bias. Previous work presents two desiderata to handle confounding bias. On the treatment group level, we aim to balance the distributions of confounder representations. On the individual level, it is desirable to capture patterns of hidden confounders that predict treatment assignments. Existing methods show the potential of utilizing network information to handle confounding bias, but they only try to satisfy one of the two desiderata. This is because the two desiderata seem to contradict each other. When the two distributions of confounder representations are highly overlapped, then we confront the undiscriminating problem between the treated and the controlled. In this work, we formulate the two desiderata as a minimax game. We propose IGNITE that learns representations of confounders from networked observational data, which is trained by a minimax game to achieve the two desiderata. Experiments verify the efficacy of IGNITE on two datasets under various settings.

Discriminative Adversarial Domain Adaptation

Given labeled instances on a source domain and unlabeled ones on a target domain, unsupervised domain adaptation aims to learn a task classifier that can well classify target instances. Recent advances rely on domain-adversarial training of deep networks to learn domain-invariant features. However, due to an issue of mode collapse induced by the separate design of task and domain classifiers, these methods are limited in aligning the joint distributions of feature and category across domains. To overcome it, we propose a novel adversarial learning method termed Discriminative Adversarial Domain Adaptation (DADA). Based on an integrated category and domain classifier, DADA has a novel adversarial objective that encourages a mutually inhibitory relation between category and domain predictions for any input instance. We show that under practical conditions, it defines a minimax game that can promote the joint distribution alignment. Except for the traditional closed set domain adaptation, we also extend DADA for extremely challenging problem settings of partial and open set domain adaptation. Experiments show the efficacy of our proposed methods and we achieve the new state of the art for all the three settings on benchmark datasets.

Fairness for Robust Log Loss Classification

Developing classification methods with high accuracy that also avoid unfair treatment of different groups has become increasingly important for data-driven decision making in social applications. Many existing methods enforce fairness constraints on a selected classifier (e.g., logistic regression) by directly forming constrained optimizations. We instead re-derive a new classifier from the first principles of distributional robustness that incorporates fairness criteria into a worst-case logarithmic loss minimization. This construction takes the form of a minimax game and produces a parametric exponential family conditional distribution that resembles truncated logistic regression. We present the theoretical benefits of our approach in terms of its convexity and asymptotic convergence. We then demonstrate the practical advantages of our approach on three benchmark fairness datasets.

Image-to-Image Translation Using Identical-Pair Adversarial Networks

We propose Identical-pair Adversarial Networks (iPANs) to solve image-to-image translation problems, such as aerial-to-map, edge-to-photo, de-raining, and night-to-daytime. Our iPANs rely mainly on the effectiveness of adversarial loss function and its network architectures. Our iPANs consist of two main networks, an image transformation network T and a discriminative network D. We use U-NET for the transformation network T and a perceptual similarity network, which has two streams of VGG16 that share the same weights for network D. Our proposed adversarial losses play a minimax game against each other based on a real identical-pair and a fake identical-pair distinguished by the discriminative network D; e.g. a discriminative network D considers two inputs as a real pair only when they are identical, otherwise a fake pair. Meanwhile, the transformation network T tries to persuade the discriminator network D that the fake pair is a real pair. We experimented on several problems of image-to-image translation and achieved results that are comparable to those of some existing approaches, such as pix2pix, and PAN.