A HYBRID ALGORITHM FOR FORMING THE SHORTEST TRAJECTORY BASED ON THE APPLICATION OF MULTI-AGENT LEARNING WITH REINFORCEMENT, THE SEARCH ALGORITHM A* AND EXCHANGE OF EXPERIENCE

The article presents a hybrid algorithm for the formation of the shortest trajectory for intelligent agents of a multi-agent system, based on the synthesis of methods of the reinforcement learning paradigm, the heuristic search algorithm A*, which has the functions of exchange of experience, as well as the automatic formation of subgroups of agents based on their visibility areas. The experimental evaluation of the developed algorithm was carried out by simulating the task of finding the target state in the maze in the Microsoft Unity environment. The results of the experiment showed that the use of the developed hybrid algorithm made it possible to reduce the time for solving the problem by an average of 12.7 % in comparison with analogs. The differences between the proposed new “hybrid algorithm for the formation of the shortest trajectory based on the use of multi-agent reinforcement learning, search algorithm A* and exchange of experience” from analogs are as follows: – application of the algorithm for the formation of subgroups of subordinate agents based on the “scope” of the leader agent for the implementation of a multi-level hierarchical system for managing a group of agents; – combining the principles of reinforcement learning and the search algorithm A*.

Deep Reinforcement Learning Versus Evolution Strategies: A Comparative Survey

<div>Deep Reinforcement Learning (DRL) and Evolution Strategies (ESs) have surpassed human-level control in many sequential decision-making problems, yet many open challenges still exist.</div><div>To get insights into the strengths and weaknesses of DRL versus ESs, an analysis of their respective capabilities and limitations is provided. </div><div>After presenting their fundamental concepts and algorithms, a comparison is provided on key aspects such as scalability, exploration, adaptation to dynamic environments, and multi-agent learning. </div><div>Then, the benefits of hybrid algorithms that combine concepts from DRL and ESs are highlighted. </div><div>Finally, to have an indication about how they compare in real-world applications, a survey of the literature for the set of applications they support is provided.</div>

Improving Multi-agent Coordination by Learning to Estimate Contention

We present a multi-agent learning algorithm, ALMA-Learning, for efficient and fair allocations in large-scale systems. We circumvent the traditional pitfalls of multi-agent learning (e.g., the moving target problem, the curse of dimensionality, or the need for mutually consistent actions) by relying on the ALMA heuristic as a coordination mechanism for each stage game. ALMA-Learning is decentralized, observes only own action/reward pairs, requires no inter-agent communication, and achieves near-optimal (<5% loss) and fair coordination in a variety of synthetic scenarios and a real-world meeting scheduling problem. The lightweight nature and fast learning constitute ALMA-Learning ideal for on-device deployment.

Too Many cooks: Bayesian inference for coordinating Multi-agent Collaboration

Collaboration requires agents to coordinate their behavior on the fly, sometimes cooperating to solve a single task together and other times dividing it up into sub-tasks to work on in parallel. Here, we develop Bayesian Delegation, a decentralized multi-agent learning mechanism with these abilities. Bayesian Delegation enables agents to rapidly infer the hidden intentions of others by inverse planning. We test Bayesian Delegation in a suite of multi-agent Markov decision processes inspired by cooking problems. On these tasks, agents with Bayesian Delegation coordinate both their high-level plans (e.g. what sub-task they should work on) and their low-level actions (e.g. avoiding getting in each other’s way). In a self-play evaluation, Bayesian Delegation outperforms alternative algorithms. Bayesian Delegation is also a capable ad-hoc collaborator and successfully coordinates with other agent types even in the absence of prior experience.

Dynamical systems as a level of cognitive analysis of multi-agent learning

A dynamical systems perspective on multi-agent learning, based on the link between evolutionary game theory and reinforcement learning, provides an improved, qualitative understanding of the emerging collective learning dynamics. However, confusion exists with respect to how this dynamical systems account of multi-agent learning should be interpreted. In this article, I propose to embed the dynamical systems description of multi-agent learning into different abstraction levels of cognitive analysis. The purpose of this work is to make the connections between these levels explicit in order to gain improved insight into multi-agent learning. I demonstrate the usefulness of this framework with the general and widespread class of temporal-difference reinforcement learning. I find that its deterministic dynamical systems description follows a minimum free-energy principle and unifies a boundedly rational account of game theory with decision-making under uncertainty. I then propose an on-line sample-batch temporal-difference algorithm which is characterized by the combination of applying a memory-batch and separated state-action value estimation. I find that this algorithm serves as a micro-foundation of the deterministic learning equations by showing that its learning trajectories approach the ones of the deterministic learning equations under large batch sizes. Ultimately, this framework of embedding a dynamical systems description into different abstraction levels gives guidance on how to unleash the full potential of the dynamical systems approach to multi-agent learning.