The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions
This work addresses the problem of developing more robust and exploratory multi-agent reinforcement learning algorithms for complex, real-world-like tasks, though it is incremental as it builds on existing benchmarks.
The paper introduces StarCraft Multi-Agent Challenges+ (SMAC+), a benchmark for multi-agent reinforcement learning that requires agents to learn multi-stage tasks and environmental factors without precise reward functions, and finds that recent algorithms perform well in familiar settings but struggle in offensive scenarios, with enhanced exploration improving but not fully solving the challenges.
In this paper, we propose a novel benchmark called the StarCraft Multi-Agent Challenges+, where agents learn to perform multi-stage tasks and to use environmental factors without precise reward functions. The previous challenges (SMAC) recognized as a standard benchmark of Multi-Agent Reinforcement Learning are mainly concerned with ensuring that all agents cooperatively eliminate approaching adversaries only through fine manipulation with obvious reward functions. This challenge, on the other hand, is interested in the exploration capability of MARL algorithms to efficiently learn implicit multi-stage tasks and environmental factors as well as micro-control. This study covers both offensive and defensive scenarios. In the offensive scenarios, agents must learn to first find opponents and then eliminate them. The defensive scenarios require agents to use topographic features. For example, agents need to position themselves behind protective structures to make it harder for enemies to attack. We investigate MARL algorithms under SMAC+ and observe that recent approaches work well in similar settings to the previous challenges, but misbehave in offensive scenarios. Additionally, we observe that an enhanced exploration approach has a positive effect on performance but is not able to completely solve all scenarios. This study proposes new directions for future research.