From Few to More: Large-scale Dynamic Multiagent Curriculum Learning

#artificialintelligence

A lot of efforts have been devoted to investigating how agents can learn effectively and achieve coordination in multiagent systems. However, it is still challenging in large-scale multiagent settings due to the complex dynamics between the environment and agents and the explosion of state-action space. In this paper, we design a novel Dynamic Multiagent Curriculum Learning (DyMA-CL) to solve large-scale problems by starting from learning on a multiagent scenario with a small size and progressively increasing the number of agents. We propose three transfer mechanisms across curricula to accelerate the learning process. Moreover, due to the fact that the state dimension varies across curricula,, and existing network structures cannot be applied in such a transfer setting since their network input sizes are fixed.


From Few to More: Large-scale Dynamic Multiagent Curriculum Learning

#artificialintelligence

A lot of efforts have been devoted to investigating how agents can learn effectively and achieve coordination in multiagent systems. However, it is still challenging in large-scale multiagent settings due to the complex dynamics between the environment and agents and the explosion of state-action space. In this paper, we design a novel Dynamic Multiagent Curriculum Learning (DyMA-CL) to solve large-scale problems by starting from learning on a multiagent scenario with a small size and progressively increasing the number of agents. We propose three transfer mechanisms across curricula to accelerate the learning process. Moreover, due to the fact that the state dimension varies across curricula,, and existing network structures cannot be applied in such a transfer setting since their network input sizes are fixed.


Preface

AAAI Conferences

Multiagent Systems (MAS) have become an important sub-field of AI, and several classical AI topics are now broadly studied in their MAS (i.e. Multiagent Planning (MAP) extends classical AI Planning to domains where several agents can plan and act together. Application areas of MAP include multi-robot environments, cooperating Internet agents, logistics, manufacturing, military tasks etc. While related MAS disciplines (e.g. Distributed Constraint Satisfaction) have benefited from standardized problem specifications and benchmarks, existing work on MAP is still very heterogeneous.


Preface

AAAI Conferences

The main goals of the workshop are as follows: (1) Familiarize researchers from different areas with the vary- ing contributions on this problem. (2) Standardize terminology and develop a taxonomy for different variants. (3) Present the state-of-the-art and discuss open challenges. (4) Encourage collaboration between participants.


A Study of AI Population Dynamics with Million-agent Reinforcement Learning

arXiv.org Artificial Intelligence

We conduct an empirical study on discovering the ordered collective dynamics obtained by a population of intelligence agents, driven by million-agent reinforcement learning. Our intention is to put intelligent agents into a simulated natural context and verify if the principles developed in the real world could also be used in understanding an artificially-created intelligent population. To achieve this, we simulate a large-scale predator-prey world, where the laws of the world are designed by only the findings or logical equivalence that have been discovered in nature. We endow the agents with the intelligence based on deep reinforcement learning (DRL). In order to scale the population size up to millions agents, a large-scale DRL training platform with redesigned experience buffer is proposed. Our results show that the population dynamics of AI agents, driven only by each agent's individual self-interest, reveals an ordered pattern that is similar to the Lotka-Volterra model studied in population biology. We further discover the emergent behaviors of collective adaptations in studying how the agents' grouping behaviors will change with the environmental resources. Both of the two findings could be explained by the self-organization theory in nature.