Search

In the first issue of Nature 2015, Google DeepMind published a paper “Human-level control through deep reinforcement learning.” Furthermore, in the first issue of Nature 2016, it published a cover paper “Mastering the game of Go with deep neural networks and tree search” and proposed the computer Go program, AlphaGo. In March 2016, AlphaGo beat the world’s top Go player Lee Sedol by 4:1. This becomes a new milestone in artificial intelligence history, the core of which is the algorithm of deep reinforcement learning (RL). [ABSTRACT FROM AUTHOR]

This paper proposes a hierarchical decision-making and control algorithm for the shepherd game, the seventh mission in the International Aerial Robotics Competition (IARC). In this game, the agent (a multirotor aerial robot) is required to contact targets (ground vehicles) sequentially and drive them to a certain boundary to earn score. During the game of 10 min, the agent should be fully autonomous without any human interference. Regarding the lower-level controller and dynamics of the agent, each action takes a duration of time to accomplish. Denoted as an action delay, in this paper, this action duration is nonconstant and is related to the final reward. Therefore, the challenging point is making the agent “aware of time” when applying a certain action. We solve this problem by two approaches: deep Q-networks and lookup table. The action delay predictor in the decision-level is fitted by a lower-level controller. Through simulations by the example of the shepherd game, the effectiveness and efficiency of this approach are validated. This paper helps our team winning the first prize in IARC 2017, and keeps the best record of this mission since it was released in 2013. [ABSTRACT FROM AUTHOR]

In spam and malware detection, attackers exploit randomization to obfuscate malicious data and increase their chances of evading detection at test time, e.g., malware code is typically obfuscated using random strings or byte sequences to hide known exploits. Interestingly, randomization has also been proposed to improve security of learning algorithms against evasion attacks, as it results in hiding information about the classifier to the attacker. Recent work has proposed game-theoretical formulations to learn secure classifiers, by simulating different evasion attacks and modifying the classification function accordingly. However, both the classification function and the simulated data manipulations have been modeled in a deterministic manner, without accounting for any form of randomization. In this paper, we overcome this limitation by proposing a randomized prediction game, namely, a noncooperative game-theoretic formulation in which the classifier and the attacker make randomized strategy selections according to some probability distribution defined over the respective strategy set. We show that our approach allows one to improve the tradeoff between attack detection and false alarms with respect to the state-of-the-art secure classifiers, even against attacks that are different from those hypothesized during design, on application examples including handwritten digit recognition, spam, and malware detection. [ABSTRACT FROM PUBLISHER]

Due to the lack of enough generalization in the state space, common methods of reinforcement learning suffer from slow learning speed, especially in the early learning trials. This paper introduces a model-based method in discrete state spaces for increasing the learning speed in terms of required experiences (but not required computation time) by exploiting generalization in the experiences of the subspaces. A subspace is formed by choosing a subset of features in the original state representation. Generalization and faster learning in a subspace are due to many-to-one mapping of experiences from the state space to each state in the subspace. Nevertheless, due to inherent perceptual aliasing (PA) in the subspaces, the policy suggested by each subspace does not generally converge to the optimal policy. Our approach, called model-based learning with subspaces (MoBLeSs), calculates the confidence intervals of the estimated $Q$ -values in the state space and in the subspaces. These confidence intervals are used in the decision-making, such that the agent benefits the most from the possible generalization while avoiding from the detriment of the PA in the subspaces. The convergence of MoBLeS to the optimal policy is theoretically investigated. In addition, we show through several experiments that MoBLeS improves the learning speed in the early trials. [ABSTRACT FROM AUTHOR]

Distributed support vector machines (DSVMs) have been developed to solve large-scale classification problems in networked systems with a large number of sensors and control units. However, the systems become more vulnerable, as detection and defense are increasingly difficult and expensive. This paper aims to develop secure and resilient DSVM algorithms under adversarial environments in which an attacker can manipulate the training data to achieve his objective. We establish a game-theoretic framework to capture the conflicting interests between an adversary and a set of distributed data processing units. The Nash equilibrium of the game allows predicting the outcome of learning algorithms in adversarial environments and enhancing the resilience of the machine learning through dynamic distributed learning algorithms. We prove that the convergence of the distributed algorithm is guaranteed without assumptions on the training data or network topologies. Numerical experiments are conducted to corroborate the results. We show that the network topology plays an important role in the security of DSVM. Networks with fewer nodes and higher average degrees are more secure. Moreover, a balanced network is found to be less vulnerable to attacks. [ABSTRACT FROM AUTHOR]

Multiagent systems (MASs) are usually applied with agents classified into leaders and followers, where selecting appropriate leaders is an important issue for formation control applications. In this paper, we investigate two leader selection problems in second-order MAS, namely, the problem of choosing up to a given number of leaders to minimize the formation error and the problem of choosing the minimum number of leaders to achieve a tolerated level of error. We propose a game theoretical method to address them. Specifically, we design a supermodular game for the leader selection problems and theoretically prove its supermodularity. In order to reach Nash equilibrium of the game, we propose strategies for the agents to learn to select leaders based on stochastic fictitious play. Extensive simulation results demonstrate that our method outperforms existing ones. [ABSTRACT FROM AUTHOR]

Regarding two-player zero-sum games of continuous-time nonlinear systems with completely unknown dynamics, this paper presents an online adaptive algorithm for learning the Nash equilibrium solution, i.e., the optimal policy pair. First, for known systems, the simultaneous policy updating algorithm (SPUA) is reviewed. A new analytical method to prove the convergence is presented. Then, based on the SPUA, without using a priori knowledge of any system dynamics, an online algorithm is proposed to simultaneously learn in real time either the minimal nonnegative solution of the Hamilton–Jacobi–Isaacs (HJI) equation or the generalized algebraic Riccati equation for linear systems as a special case, along with the optimal policy pair. The approximate solution to the HJI equation and the admissible policy pair is reexpressed by the approximation theorem. The unknown constants or weights of each are identified simultaneously by resorting to the recursive least square method. The convergence of the online algorithm to the optimal solutions is provided. A practical online algorithm is also developed. Simulation results illustrate the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

This paper aims to investigate the fuzzy constrained matrix game (MG) problems using the concepts of recurrent neural networks (RNNs). To the best of our knowledge, this paper is the first in attempting to find a solution for fuzzy game problems using RNN models. For this purpose, a fuzzy game problem is reformulated into a weighting problem. Then, the Karush–Kuhn–Tucker (KKT) optimality conditions are provided for the weighting problem. The KKT conditions are used to propose the RNN model. Moreover, the Lyapunov stability and the global convergence of the RNN model are also confirmed. Finally, three illustrative examples are presented to demonstrate the effectiveness of this approach. The obtained results are compared with the results obtained by the previous approaches for solving fuzzy constrained MG. [ABSTRACT FROM AUTHOR]

In this paper, approximate optimal distributed control schemes for a class of nonlinear interconnected systems with strong interconnections are presented using continuous and event-sampled feedback information. The optimal control design is formulated as an $N$ -player nonzero-sum game where the control policies of the subsystems act as players. An approximate Nash equilibrium solution to the game, which is the solution to the coupled Hamilton–Jacobi equation, is obtained using the approximate dynamic programming-based approach. A critic neural network (NN) at each subsystem is utilized to approximate the Nash solution and novel event-sampling conditions, that are decentralized, are designed to asynchronously orchestrate the sampling and transmission of state vector at each subsystem. To ensure the local ultimate boundedness of the closed-loop system state and NN parameter estimation errors, a hybrid-learning scheme is introduced and the stability is guaranteed using Lyapunov-based stability analysis. Finally, implementation of the proposed event-based distributed control scheme for linear interconnected systems is discussed. For completeness, Zeno-free behavior of the event-sampled system is shown analytically and a numerical example is included to support the analytical results. [ABSTRACT FROM AUTHOR]

In this paper, two aggregative games over weight-balanced digraphs are studied, where the cost functions of all players depend on not only their own decisions but also the aggregate of all decisions. In the first problem, the cost functions of players are differentiable with Lipschitz gradients, and the decisions of all players are coupled by linear coupling constraints. In the second problem, the cost functions are nonsmooth, and the decisions of all players are constrained by local feasibility constraints as well as linear coupling constraints. In order to seek the variational generalized Nash equilibrium (GNE) of the differentiable aggregative games, a continuous-time distributed algorithm is developed via gradient descent and dynamic average consensus, and its exponential convergence to the variational GNE is proven with the help of Lyapunov stability theory. Then, another continuous-time distributed projection-based algorithm is proposed for the nonsmooth aggregative games based on differential inclusions and differentiated projection operations. Moreover, the convergence of the algorithm to the variational GNE is analyzed by utilizing singular perturbation analysis. Finally, simulation examples are presented to illustrate the effectiveness of our methods. [ABSTRACT FROM AUTHOR]

The decision-making process, which is regarded as cognitive and ubiquitous, has been exploited in diverse fields, such as psychology, economics, and artificial intelligence. This paper considers the problem of modeling agent cognition in a class of game-theoretic decision-making scenarios called extensive games. We present a novel framework in which artificial neural networks are incorporated to simulate agent cognition regarding the structure of the underlying game and the goodness of the game situations therein. An algorithmic procedure is investigated to describe the process for solving games with cognition, and then, a new equilibrium concept is proposed as a refinement of the classical one—subgame perfect equilibrium—by involving players’ cognitive reasoning. Moreover, a series of results concerning the computational complexity, soundness, and completeness of the algorithm, as well as the existence of an equilibrium solution, is obtained. This framework, which is shown to be general enough to model the way in which AlphaGo plays Go, may offer a means for bridging the gap between theoretical models and practical problem-solving. [ABSTRACT FROM AUTHOR]

Reinforcement learning in environments with many action–state pairs is challenging. The issue is the number of episodes needed to thoroughly search the policy space. Most conventional heuristics address this search problem in a stochastic manner. This can leave large portions of the policy space unvisited during the early training stages. In this paper, we propose an uncertainty-based, information-theoretic approach for performing guided stochastic searches that more effectively cover the policy space. Our approach is based on the value of information, a criterion that provides the optimal tradeoff between expected costs and the granularity of the search process. The value of information yields a stochastic routine for choosing actions during learning that can explore the policy space in a coarse to fine manner. We augment this criterion with a state-transition uncertainty factor, which guides the search process into previously unexplored regions of the policy space. We evaluate the uncertainty-based value-of-information policies on the games Centipede and Crossy Road. Our results indicate that our approach yields better performing policies in fewer episodes than stochastic-based exploration strategies. We show that the training rate for our approach can be further improved by using the policy cross entropy to guide our criterion’s hyperparameter selection. [ABSTRACT FROM AUTHOR]

In this paper, a new training paradigm is proposed for deep reinforcement learning using self-paced prioritized curriculum learning with coverage penalty. The proposed deep curriculum reinforcement learning (DCRL) takes the most advantage of experience replay by adaptively selecting appropriate transitions from replay memory based on the complexity of each transition. The criteria of complexity in DCRL consist of self-paced priority as well as coverage penalty. The self-paced priority reflects the relationship between the temporal-difference error and the difficulty of the current curriculum for sample efficiency. The coverage penalty is taken into account for sample diversity. With comparison to deep Q network (DQN) and prioritized experience replay (PER) methods, the DCRL algorithm is evaluated on Atari 2600 games, and the experimental results show that DCRL outperforms DQN and PER on most of these games. More results further show that the proposed curriculum training paradigm of DCRL is also applicable and effective for other memory-based deep reinforcement learning approaches, such as double DQN and dueling network. All the experimental results demonstrate that DCRL can achieve improved training efficiency and robustness for deep reinforcement learning. [ABSTRACT FROM AUTHOR]

In this paper, based on the adaptive critic learning technique, the H\infty control for a class of unknown nonlinear dynamic systems is investigated by adopting a mixed data and event driven design approach. The nonlinear H\infty control problem is formulated as a two-player zero-sum differential game and the adaptive critic method is employed to cope with the data-based optimization. The novelty lies in that the data driven learning identifier is combined with the event driven design formulation, in order to develop the adaptive critic controller, thereby accomplishing the nonlinear H\infty control. The event driven optimal control law and the time driven worst case disturbance law are approximated by constructing and tuning a critic neural network. Applying the event driven feedback control, the closed-loop system is built with stability analysis. Simulation studies are conducted to verify the theoretical results and illustrate the control performance. It is significant to observe that the present research provides a new avenue of integrating data-based control and event-triggering mechanism into establishing advanced adaptive critic systems. [ABSTRACT FROM AUTHOR]

This paper establishes an off-policy integral reinforcement learning (IRL) method to solve nonlinear continuous-time (CT) nonzero-sum (NZS) games with unknown system dynamics. The IRL algorithm is presented to obtain the iterative control and off-policy learning is used to allow the dynamics to be completely unknown. Off-policy IRL is designed to do policy evaluation and policy improvement in the policy iteration algorithm. Critic and action networks are used to obtain the performance index and control for each player. The gradient descent algorithm makes the update of critic and action weights simultaneously. The convergence analysis of the weights is given. The asymptotic stability of the closed-loop system and the existence of Nash equilibrium are proved. The simulation study demonstrates the effectiveness of the developed method for nonlinear CT NZS games with unknown system dynamics. [ABSTRACT FROM PUBLISHER]

Pairwise-interaction graphical games have been widely used in the study and design of strategic interaction in multiagent systems. With regard to this issue, one entitative problem is actually to understand how the interaction structure of agents affects the strategy configuration of Nash equilibria. This paper intends to study the effect of interaction networks on Nash equilibria in pairwise-interaction graphical games. We first show that interaction networks may induce new strategy equilibria in pairwise-interaction graphical games and then provide graphical conditions for the existence of these network-induced equilibria. Furthermore, to determine Nash equilibria of pairwise-interaction graphical games, a graphical replicator dynamics model is formulated, and its connection with graphical games is established. In detail, it is shown that every Nash equilibrium of the graphical games corresponds to a fixed point of the graphical replicator dynamics and that every asymptotically stable fixed point of the graphical replicator dynamics corresponds to a strict pure Nash equilibrium of the graphical games. The obtained results are applied in understanding coordination in complex networks and determination of structural conflicts in signed graphs. This work may provide new insights into understanding and designing strategy equilibria and dynamics in games on networks. [ABSTRACT FROM AUTHOR]

This paper provides necessary and sufficient conditions for the existence of the static output-feedback (OPFB) solution to the H∞ control problem for linear discrete-time systems. It is shown that the solution of the static OPFB H∞ control is a Nash equilibrium point. Furthermore, a Q-learning algorithm is developed to find the H∞ OPFB solution online using data measured along the system trajectories and without knowing the system matrices. This is achieved by solving a game algebraic Riccati equation online and using the measured data. A simulation example shows the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Existing smart grid security research investigates different attack techniques and cascading failures from the attackers’ viewpoints, while the defenders’ or the operators’ protection strategies are somehow neglected. Game theoretic methods are applied for the attacker–defender games in the smart grid security area. Yet, most of the existing works only use the one-shot game and do not consider the dynamic process of the electric power grid. In this paper, we propose a new solution for a multistage game (also called a dynamic game) between the attacker and the defender based on reinforcement learning to identify the optimal attack sequences given certain objectives (e.g., transmission line outages or generation loss). Different from a one-shot game, the attacker here learns a sequence of attack actions applying for the transmission lines and the defender protects a set of selected lines. After each time step, the cascading failure will be measured, and the line outage (and/or generation loss) will be used as the feedback for the attacker to generate the next action. The performance is evaluated on W&W 6-bus and IEEE 39-bus systems. A comparison between a multistage attack and a one-shot attack is conducted to show the significance of the multistage attack. Furthermore, different protection strategies are evaluated in simulation, which shows that the proposed reinforcement learning solution can identify optimal attack sequences under several attack objectives. It also indicates that attacker’s learned information helps the defender to enhance the security of the system. [ABSTRACT FROM AUTHOR]

Social dilemmas have attracted extensive interest in the research of multiagent systems in order to study the emergence of cooperative behaviors among selfish agents. Understanding how agents can achieve cooperation in social dilemmas through learning from local experience is a critical problem that has motivated researchers for decades. This paper investigates the possibility of exploiting emotions in agent learning in order to facilitate the emergence of cooperation in social dilemmas. In particular, the spatial version of social dilemmas is considered to study the impact of local interactions on the emergence of cooperation in the whole system. A double-layered emotional multiagent reinforcement learning framework is proposed to endow agents with internal cognitive and emotional capabilities that can drive these agents to learn cooperative behaviors. Experimental results reveal that various network topologies and agent heterogeneities have significant impacts on agent learning behaviors in the proposed framework, and under certain circumstances, high levels of cooperation can be achieved among the agents. [ABSTRACT FROM PUBLISHER]

Kmeans-type clustering aims at partitioning a data set into clusters such that the objects in a cluster are compact and the objects in different clusters are well separated. However, most kmeans-type clustering algorithms rely on only intracluster compactness while overlooking intercluster separation. In this paper, a series of new clustering algorithms by extending the existing kmeans-type algorithms is proposed by integrating both intracluster compactness and intercluster separation. First, a set of new objective functions for clustering is developed. Based on these objective functions, the corresponding updating rules for the algorithms are then derived analytically. The properties and performances of these algorithms are investigated on several synthetic and real-life data sets. Experimental studies demonstrate that our proposed algorithms outperform the state-of-the-art kmeans-type clustering algorithms with respect to four metrics: accuracy, RandIndex, Fscore, and normal mutual information. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (RL) comprehensively uses the psychological mechanisms of “trial and error” and “reward and punishment” in RL as well as powerful feature expression and nonlinear mapping in deep learning. Currently, it plays an essential role in the fields of artificial intelligence and machine learning. Since an RL agent needs to constantly interact with its surroundings, the deep Q network (DQN) is inevitably faced with the need to learn numerous network parameters, which results in low learning efficiency. In this paper, a multisource transfer double DQN (MTDDQN) based on actor learning is proposed. The transfer learning technique is integrated with deep RL to make the RL agent collect, summarize, and transfer action knowledge, including policy mimic and feature regression, to the training of related tasks. There exists action overestimation in DQN, i.e., the lower probability limit of action corresponding to the maximum Q value is nonzero. Therefore, the transfer network is trained by using double DQN to eliminate the error accumulation caused by action overestimation. In addition, to avoid negative transfer, i.e., to ensure strong correlations between source and target tasks, a multisource transfer learning mechanism is applied. The Atari2600 game is tested on the arcade learning environment platform to evaluate the feasibility and performance of MTDDQN by comparing it with some mainstream approaches, such as DQN and double DQN. Experiments prove that MTDDQN achieves not only human-like actor learning transfer capability, but also the desired learning efficiency and testing accuracy on target task. [ABSTRACT FROM AUTHOR]

This paper reviews the current state of the art on reinforcement learning (RL)-based feedback control solutions to optimal regulation and tracking of single and multiagent systems. Existing RL solutions to both optimal \mathcal H2 and \mathcal H_\infty control problems, as well as graphical games, will be reviewed. RL methods learn the solution to optimal control and game problems online and using measured data along the system trajectories. We discuss Q-learning and the integral RL algorithm as core algorithms for discrete-time (DT) and continuous-time (CT) systems, respectively. Moreover, we discuss a new direction of off-policy RL for both CT and DT systems. Finally, we review several applications. [ABSTRACT FROM AUTHOR]

This paper develops an off-policy reinforcement learning (RL) algorithm to solve optimal synchronization of multiagent systems. This is accomplished by using the framework of graphical games. In contrast to traditional control protocols, which require complete knowledge of agent dynamics, the proposed off-policy RL algorithm is a model-free approach, in that it solves the optimal synchronization problem without knowing any knowledge of the agent dynamics. A prescribed control policy, called behavior policy, is applied to each agent to generate and collect data for learning. An off-policy Bellman equation is derived for each agent to learn the value function for the policy under evaluation, called target policy, and find an improved policy, simultaneously. Actor and critic neural networks along with least-square approach are employed to approximate target control policies and value functions using the data generated by applying prescribed behavior policies. Finally, an off-policy RL algorithm is presented that is implemented in real time and gives the approximate optimal control policy for each agent using only measured data. It is shown that the optimal distributed policies found by the proposed algorithm satisfy the global Nash equilibrium and synchronize all agents to the leader. Simulation results illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

$H_\infty $ control is a powerful method to solve the disturbance attenuation problems that occur in some control systems. The design of such controllers relies on solving the zero-sum game (ZSG). But in practical applications, the exact dynamics is mostly unknown. Identification of dynamics also produces errors that are detrimental to the control performance. To overcome this problem, an iterative adaptive dynamic programming algorithm is proposed in this paper to solve the continuous-time, unknown nonlinear ZSG with only online data. A model-free approach to the Hamilton–Jacobi–Isaacs equation is developed based on the policy iteration method. Control and disturbance policies and value are approximated by neural networks (NNs) under the critic–actor–disturber structure. The NN weights are solved by the least-squares method. According to the theoretical analysis, our algorithm is equivalent to a Gauss–Newton method solving an optimization problem, and it converges uniformly to the optimal solution. The online data can also be used repeatedly, which is highly efficient. Simulation results demonstrate its feasibility to solve the unknown nonlinear ZSG. When compared with other algorithms, it saves a significant amount of online measurement time. [ABSTRACT FROM PUBLISHER]

Hyperspectral (HS) pansharpening is of great importance in improving the spatial resolution of HS images for remote sensing tasks. HS image comprises abundant spectral contents, whereas panchromatic (PAN) image provides spatial information. HS pansharpening constitutes the possibility for providing the pansharpened image with both high spatial and spectral resolution. This article develops a specific pansharpening framework based on a generative dual-adversarial network (called PS-GDANet). Specifically, the pansharpening problem is formulated as a dual task that can be solved by a generative adversarial network (GAN) with two discriminators. The spatial discriminator forces the intensity component of the pansharpened image to be as consistent as possible with the PAN image, and the spectral discriminator helps to preserve spectral information of the original HS image. Instead of designing a deep network, PS-GDANet extends GANs to two discriminators and provides a high-resolution pansharpened image in a fraction of iterations. The experimental results demonstrate that PS-GDANet outperforms several widely accepted state-of-the-art pansharpening methods in terms of qualitative and quantitative assessment. [ABSTRACT FROM AUTHOR]

We study the decentralized event-driven control problem of nonlinear dynamical systems with mismatched interconnections and asymmetric input constraints. To begin with, by introducing a discounted cost function for each auxiliary subsystem, we transform the decentralized event-driven constrained control problem into a group of nonlinear $H_{2}$ -constrained optimal control problems. Then, we develop the event-driven Hamilton–Jacobi–Bellman equations (ED-HJBEs), which arise in the nonlinear $H_{2}$ -constrained optimal control problems. Meanwhile, we demonstrate that all the solutions of the ED-HJBEs together keep the overall system stable in the sense of uniform ultimate boundedness (UUB). To solve the ED-HJBEs, we build a critic-only architecture under the framework of adaptive critic designs. The architecture only employs critic neural networks and updates their weight vectors via the gradient descent method. After that, based on the Lyapunov approach, we prove that the UUB stability of all signals in the closed-loop auxiliary subsystems is assured. Finally, simulations of an illustrated nonlinear interconnected plant are provided to validate the present designs. [ABSTRACT FROM AUTHOR]

For nonzero-sum (NZS) games of nonlinear systems, reinforcement learning (RL) or adaptive dynamic programming (ADP) has shown its capability of approximating the desired index performance and the optimal input policy iteratively. In this article, an event-triggered ADP is proposed for NZS games of continuous-time nonlinear systems with completely unknown system dynamics. To achieve the Nash equilibrium solution approximately, the critic neural networks and actor neural networks are utilized to estimate the value functions and the control policies, respectively. Compared with the traditional time-triggered mechanism, the proposed algorithm updates the neural network weights as well as the inputs of players only when a state-based event-triggered condition is violated. It is shown that the system stability and the weights’ convergence are still guaranteed under mild assumptions, while occupation of communication and computation resources is considerably reduced. Meanwhile, the infamous Zeno behavior is excluded by proving the existence of a minimum inter-event time (MIET) to ensure the feasibility of the closed-loop event-triggered continuous-time system. Finally, a numerical example is simulated to illustrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

The Nash equilibrium is an important concept in game theory. It describes the least exploitability of one player from any opponents. We combine game theory, dynamic programming, and recent deep reinforcement learning (DRL) techniques to online learn the Nash equilibrium policy for two-player zero-sum Markov games (TZMGs). The problem is first formulated as a Bellman minimax equation, and generalized policy iteration (GPI) provides a double-loop iterative way to find the equilibrium. Then, neural networks are introduced to approximate Q functions for large-scale problems. An online minimax Q network learning algorithm is proposed to train the network with observations. Experience replay, dueling network, and double Q-learning are applied to improve the learning process. The contributions are twofold: 1) DRL techniques are combined with GPI to find the TZMG Nash equilibrium for the first time and 2) the convergence of the online learning algorithm with a lookup table and experience replay is proven, whose proof is not only useful for TZMGs but also instructive for single-agent Markov decision problems. Experiments on different examples validate the effectiveness of the proposed algorithm on TZMG problems. [ABSTRACT FROM AUTHOR]

We present JueWu-SL, the first supervised-learning-based artificial intelligence (AI) program that achieves human-level performance in playing multiplayer online battle arena (MOBA) games. Unlike prior attempts, we integrate the macro-strategy and the micromanagement of MOBA-game-playing into neural networks in a supervised and end-to-end manner. Tested on Honor of Kings, the most popular MOBA at present, our AI performs competitively at the level of High King players in standard 5v5 games. [ABSTRACT FROM AUTHOR]

The evolutionarily stable strategy (ESS) of networked evolutionary games (NEGs) is studied. Analyzing the ESS of infinite popular evolutionary games and comparing it with networked games, a new verifiable definition of ESS for NEGs is proposed. Then, the fundamental evolutionary equation (FEE) is investigated and used to construct the strategy profile dynamics (SPDs) of homogeneous NEGs. Two ways for verifying the ESS are proposed: 1) using the SPDs to verify it directly. The SPDs provides complete information about the NEGs, and then necessary and sufficient conditions are revealed. It can be used for NEGs with small size and 2) some sufficient conditions are proposed to verify the ESS of NEGs via their FEEs. This method is particularly suitable for large scale networks. Some illustrative examples are included to demonstrate the theoretical results. [ABSTRACT FROM AUTHOR]

In this article, an online adaptive optimal control algorithm based on adaptive dynamic programming is developed to solve the multiplayer nonzero-sum game (MP-NZSG) for discrete-time unknown nonlinear systems. First, a model-free coupled globalized dual-heuristic dynamic programming (GDHP) structure is designed to solve the MP-NZSG problem, in which there is no model network or identifier. Second, in order to relax the requirement of systems dynamics, an online adaptive learning algorithm is developed to solve the Hamilton–Jacobi equation using the system states of two adjacent time steps. Third, a series of critic networks and action networks are used to approximate value functions and optimal policies for all players. All the neural network (NN) weights are updated online based on real-time system states. Fourth, the uniformly ultimate boundedness analysis of the NN approximation errors is proved based on the Lyapunov approach. Finally, simulation results are given to demonstrate the effectiveness of the developed scheme. [ABSTRACT FROM AUTHOR]

In this article, a novel learning methodology is introduced for the problem of classification in the context of high-dimensional data. In particular, the challenges introduced by high-dimensional data sets are addressed by formulating a $L_{1}$ regularized zero-sum game where optimal sparsity is estimated through a two-player game between the penalty coefficients/sparsity parameters and the deep neural network weights. In order to solve this game, a distributed learning methodology is proposed where additional variables are utilized to derive layerwise cost functions. Finally, an alternating minimization approach developed to solve the problem where the Nash solution provides optimal sparsity and compensation through the classifier. The proposed learning approach is implemented in a parallel and distributed environment through a novel computational algorithm. The efficiency of the approach is demonstrated both theoretically and empirically with nine data sets. [ABSTRACT FROM AUTHOR]

Considering the fact that in the real world, a certain agent may have some sort of advantage to act before others, a novel hierarchical optimal synchronization problem for linear systems, composed of one major agent and multiple minor agents, is formulated and studied in this article from a Stackelberg–Nash game perspective. The major agent herein makes its decision prior to others, and then, all the minor agents determine their actions simultaneously. To seek the optimal controllers, the Hamilton–Jacobi–Bellman (HJB) equations in coupled forms are established, whose solutions are further proven to be stable and constitute the Stackelberg–Nash equilibrium. Due to the introduction of the asymmetric roles for agents, the established HJB equations are more strongly coupled and more difficult to solve than that given in most existing works. Therefore, we propose a new reinforcement learning (RL) algorithm, i.e., a two-level value iteration (VI) algorithm, which does not rely on complete system matrices. Furthermore, the proposed algorithm is shown to be convergent, and the converged values are exactly the optimal ones. To implement this VI algorithm, neural networks (NNs) are employed to approximate the value functions, and the gradient descent method is used to update the weights of NNs. Finally, an illustrative example is provided to verify the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Control-theoretic differential games have been used to solve optimal control problems in multiplayer systems. Most existing studies on differential games either assume deterministic dynamics or dynamics corrupted with additive noise. In realistic environments, multidimensional environmental uncertainties often modulate system dynamics in a more complicated fashion. In this article, we study stochastic multiplayer differential games, where the players’ dynamics are modulated by randomly time-varying parameters. We first formulate two differential games for systems of general uncertain linear dynamics, including the two-player zero-sum and multiplayer nonzero-sum games. We then show that optimal control policies, which constitute the Nash equilibrium solutions, can be derived from the corresponding Hamiltonian functions. Stability is proven using the Lyapunov type of analysis. In order to solve the stochastic differential games online, we integrate reinforcement learning (RL) and an effective uncertainty sampling method called the multivariate probabilistic collocation method (MPCM). Two learning algorithms, including the on-policy integral RL (IRL) and off-policy IRL, are designed for the formulated games, respectively. We show that the proposed learning algorithms can effectively find the Nash equilibrium solutions for the stochastic multiplayer differential games. [ABSTRACT FROM AUTHOR]

This article studies an event-triggered communication and adaptive dynamic programming (ADP) co-design control method for the multiplayer nonzero-sum (NZS) games of a class of nonlinear discrete-time wireless sensor-actuator network (WSAN) systems subject to input constraints. By virtue of the ADP algorithm, the critic and actor networks are established to attain the approximate Nash equilibrium point solution in the context of the constrained control mechanism. Simultaneously, as the sensors and actuators are physically distributed, a decentralized event-triggered communication protocol is presented, accompanied by a dead-zone operation which avoids the unnecessary events. By predefining the triggering thresholds and compensation values, a novel adaptive triggering condition is derived to guarantee the stability of the event-based closed-loop control system. Then resorting to the Lyapunov theory, the system states and the critic/actor network weight estimation errors are proven to be ultimately bounded. Moreover, an explicit analysis on the nontriviality of the interevent times is also provided. Finally, two numerical examples are conducted to validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

We investigate the adversarial multiarmed bandit problem and introduce an online algorithm that asymptotically achieves the performance of the best switching bandit arm selection strategy. Our algorithms are truly online such that we do not use the game length or the number of switches of the best arm selection strategy in their constructions. Our results are guaranteed to hold in an individual sequence manner, since we have no statistical assumptions on the bandit arm losses. Our regret bounds, i.e., our performance bounds with respect to the best bandit arm selection strategy, are minimax optimal up to logarithmic terms. We achieve the minimax optimal regret with computational complexity only log-linear in the game length. Thus, our algorithms can be efficiently used in applications involving big data. Through an extensive set of experiments involving synthetic and real data, we demonstrate significant performance gains achieved by the proposed algorithm with respect to the state-of-the-art switching bandit algorithms. We also introduce a general efficiently implementable bandit arm selection framework, which can be adapted to various applications. [ABSTRACT FROM AUTHOR]

In this brief, an online robust adaptive dynamic programming algorithm is proposed for two-player zero-sum games of continuous-time unknown linear systems with matched uncertainties, which are functions of system outputs and states of a completely unknown exosystem. The online algorithm is developed using the policy iteration (PI) scheme with only one iteration loop. A new analytical method is proposed for convergence proof of the PI scheme. The sufficient conditions are given to guarantee globally asymptotic stability and suboptimal property of the closed-loop system. Simulation studies are conducted to illustrate the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

An approximate online equilibrium solution is developed for an $N$ -player nonzero-sum game subject to continuous-time nonlinear unknown dynamics and an infinite horizon quadratic cost. A novel actor–critic–identifier structure is used, wherein a robust dynamic neural network is used to asymptotically identify the uncertain system with additive disturbances, and a set of critic and actor NNs are used to approximate the value functions and equilibrium policies, respectively. The weight update laws for the actor neural networks (NNs) are generated using a gradient-descent method, and the critic NNs are generated by least square regression, which are both based on the modified Bellman error that is independent of the system dynamics. A Lyapunov-based stability analysis shows that uniformly ultimately bounded tracking is achieved, and a convergence analysis demonstrates that the approximate control policies converge to a neighborhood of the optimal solutions. The actor, critic, and identifier structures are implemented in real time continuously and simultaneously. Simulations on two and three player games illustrate the performance of the developed method. [ABSTRACT FROM PUBLISHER]