Laser Learning Environment: A new environment for coordination-critical multi-agent tasks
This addresses the problem of developing effective multi-agent coordination algorithms for AI researchers, though it is incremental as it focuses on benchmarking rather than solving the problem.
The authors introduced the Laser Learning Environment (LLE), a multi-agent reinforcement learning environment where coordination is critical, and showed that state-of-the-art MARL algorithms consistently fail at collaborative tasks due to their inability to escape state space bottlenecks caused by zero-incentive dynamics.
We introduce the Laser Learning Environment (LLE), a collaborative multi-agent reinforcement learning environment in which coordination is central. In LLE, agents depend on each other to make progress (interdependence), must jointly take specific sequences of actions to succeed (perfect coordination), and accomplishing those joint actions does not yield any intermediate reward (zero-incentive dynamics). The challenge of such problems lies in the difficulty of escaping state space bottlenecks caused by interdependence steps since escaping those bottlenecks is not rewarded. We test multiple state-of-the-art value-based MARL algorithms against LLE and show that they consistently fail at the collaborative task because of their inability to escape state space bottlenecks, even though they successfully achieve perfect coordination. We show that Q-learning extensions such as prioritized experience replay and n-steps return hinder exploration in environments with zero-incentive dynamics, and find that intrinsic curiosity with random network distillation is not sufficient to escape those bottlenecks. We demonstrate the need for novel methods to solve this problem and the relevance of LLE as cooperative MARL benchmark.