Multi-agent systems have a wide range of applications in cooperative and competitive tasks. As the number of agents increases, nonstationarity gets more serious in multi-agent reinforcement learning (MARL), which brings great difficulties to the learning process. Besides, current mainstream algorithms configure each agent an independent network,so that the memory usage increases linearly with the number of agents which greatly slows down the interaction with the environment. Inspired by Generative Adversarial Networks (GAN), this paper proposes an iterative update method (IU) to stabilize the nonstationary environment. Further, we add first-person perspective and represent all agents by only one network which can change agents' policies from sequential compute to batch compute. Similar to continual lifelong learning, we realize the iterative update method in this unified representative network (IUUR). In this method, iterative update can greatly alleviate the nonstationarity of the environment, unified representation can speed up the interaction with environment and avoid the linear growth of memory usage. Besides, this method does not bother decentralized execution and distributed deployment. Experiments show that compared with MADDPG, our algorithm achieves state-of-the-art performance and saves wall-clock time by a large margin especially with more agents.

In this paper, we consider the problem of collectively exploring unknown and dynamic environments with a decentralized heterogeneous multi-robot system consisting of multiple units of two variants of a miniature robot. The first variant-a wheeled ground unit-is at the core of a swarm of floor-mapping robots exhibiting scalability, robustness and flexibility. These properties are systematically tested and quantitatively evaluated in unstructured and dynamic environments, in the absence of any supporting infrastructure. The results of repeated sets of experiments show a consistent performance for all three features, as well as the possibility to inject units into the system while it is operating. Several units of the second variant-a wheg-based wall-climbing unit-are used to support the swarm of mapping robots when simultaneously exploring multiple floors by expanding the distributed communication channel necessary for the coordinated behavior among platforms. Although the occupancy-grid maps obtained can be large, they are fully distributed. Not a single robotic unit possesses the overall map, which is not required by our cooperative path-planning strategy.

The presence of artificial agents in human social networks is growing. From chatbots to robots, human experience in the developed world is moving towards a socio-technical system in which agents can be technological or biological, with increasingly blurred distinctions between. Given that emotion is a key element of human interaction, enabling artificial agents with the ability to reason about affect is a key stepping stone towards a future in which technological agents and humans can work together. This paper presents work on building intelligent computational agents that integrate both emotion and cognition. These agents are grounded in the well-established social-psychological Bayesian Affect Control Theory (BayesAct). The core idea of BayesAct is that humans are motivated in their social interactions by affective alignment: they strive for their social experiences to be coherent at a deep, emotional level with their sense of identity and general world views as constructed through culturally shared symbols. This affective alignment creates cohesive bonds between group members, and is instrumental for collaborations to solidify as relational group commitments. BayesAct agents are motivated in their social interactions by a combination of affective alignment and decision theoretic reasoning, trading the two off as a function of the uncertainty or unpredictability of the situation. This paper provides a high-level view of dual process theories and advances BayesAct as a plausible, computationally tractable model based in social-psychological theory. We introduce a revised BayesAct model that more deeply integrates social-psychological theorising, and we demonstrate a component of the model as being sufficient to account for cognitive biases about fairness, dissonance and conformity. We show how the model can unify different exploration strategies in reinforcement learning.

Understanding and reasoning about physics is an important ability of intelligent agents. We develop the PHYRE benchmark for physical reasoning that contains a set of simple classical mechanics puzzles in a 2D physical environment. The benchmark is designed to encourage the development of learning algorithms that are sample-efficient and generalize well across puzzles. We test several modern learning algorithms on PHYRE and find that these algorithms fall short in solving the puzzles efficiently. We expect that PHYRE will encourage the development of novel sample-efficient agents that learn efficient but useful models of physics. For code and to play PHYRE for yourself, please visit https://player.phyre.ai.

To cooperate with humans effectively, virtual agents need to be able to understand and execute language instructions. A typical setup to achieve this is with a scripted teacher which guides a virtual agent using language instructions. However, such setup has clear limitations in scalability and, more importantly, it is not interactive. Here, we introduce an autonomous agent that uses discrete communication to interactively guide other agents to navigate and act on a simulated environment. The developed communication protocol is trainable, emergent and requires no additional supervision. The emergent language speeds up learning of new agents, it generalizes across incrementally more difficult tasks and, contrary to most other emergent languages, it is highly interpretable. We demonstrate how the emitted messages correlate with particular actions and observations, and how new agents become less dependent on this guidance as training progresses. By exploiting the correlations identified in our analysis, we manage to successfully address the agents in their own language.

Abstract--The novel research area of computational empathy is in its infancy and moving towards developing methods and standards. One major problem is the lack of agreement on the evaluation of empathy in artificial interactive systems . Even though the existence of well-established methods from psyc hol-ogy, psychiatry and neuroscience, the translation between these methods and computational empathy is not straightforward. It requires a collective effort to develop metrics that are mor e suitable for interactive artificial agents. This paper is ai med as an attempt to initiate the dialogue on this important proble m. We examine the evaluation methods for empathy in humans and provide suggestions for the development of better metri cs to evaluate empathy in artificial agents. We acknowledge the difficulty of arriving at a single solution in a vast variety o f interactive systems and propose a set of systematic approac hes that can be used with a variety of applications and systems. Emerging technologies continue to change the ways in which we interact with computers. Computational systems are evolving from being mere tools to assistants, trainers and companion agents. All of these new roles assigned to these systems highlight the importance of embodying these agents with social and emotional capabilities.

Here's our daily update in tweets, live from IJCAI (International Joint Conference on Artificial Intelligence) in Macau. Like yesterday, we'll be covering tutorials and workshops. Now attending the #tutorial "Argumentation and Machine Learning: When the Whole is Greater than the Sum of its Parts" by @CeruttiFederico, & learning about #ML mechanisms that create, annotate, analyze & evaluate arguments expressed in natural language.#AI Now: "Dialogues with Socially Aware Robot Agents – Knowledge & Reasoning using Natural Language," an invited #IJCAI2019 talk by Prof. Kristiina Jokinen Her start: "The quality of #intelligence possessed by humans and #AI is fundamentally different."#Bridging2019 On his second slide: #AGI "needs fresh methods with cognitive architectures and philosophy of mind."#AI

Manipulation and locomotion are closely related problems that are often studied in isolation. In this work, we study the problem of coordinating multiple mobile agents to exhibit manipulation behaviors using a reinforcement learning (RL) approach. Our method hinges on the use of hierarchical sim2real -- a simulated environment is used to learn low-level goal-reaching skills, which are then used as the action space for a high-level RL controller, also trained in simulation. The full hierarchical policy is then transferred to the real world in a zero-shot fashion. The application of domain randomization during training enables the learned behaviors to generalize to real-world settings, while the use of hierarchy provides a modular paradigm for learning and transferring increasingly complex behaviors. We evaluate our method on a number of real-world tasks, including coordinated object manipulation in a multi-agent setting. See videos at https://sites.google.com/view/manipulation-via-locomotion

Abstract-- In this paper, we describe a framework for autonomous decision making in a dynamic and interactive environment based on cognitive hierarchy theory. We model the in - teractions between the ego agent and its operating environm ent as a two-player dynamic game, and integrate cognitive behav - ioral models, Bayesian inference, and receding-horizon op timal control to define a dynamically-evolving decision strategy for the ego agent. Simulation examples representing autonomou s vehicle control in three traffic scenarios where the autonom ous ego vehicle interacts with a human-driven vehicle are repor ted. Autonomous systems are becoming more capable, better accepted, and more commonplace. Many autonomous systems, including collaborative robots [1] and self-driv ing cars [2], operate in dynamic and interactive environments.

Deep Reinforcement Learning has made significant progress in multi-agent systems in recent years. In this review article, we have mostly focused on recent papers on Multi-Agent Reinforcement Learning (MARL) than the older papers, unless it was necessary. Several ideas and papers are proposed with different notations, and we tried our best to unify them with a single notation and categorize them by their relevance. In particular, we have focused on five common approaches on modeling and solving multi-agent reinforcement learning problems: (I) independent-learners, (II) fully observable critic, (III) value function decomposition, (IV) consensus, (IV) learn to communicate. Moreover, we discuss some new emerging research areas in MARL along with the relevant recent papers. In addition, some of the recent applications of MARL in real world are discussed. Finally, a list of available environments for MARL research are provided and the paper is concluded with proposals on the possible research directions.