Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents
This work tackles the problem of aligning RL agent behaviors with biological risk-avoidance for applications in robotics or AI safety, though it is incremental in bridging the gap between artificial and biological systems.
The paper compared reinforcement learning (RL) agents to biological mice in a predator-avoidance maze, finding that RL agents lack self-preservation instincts and take excessive risks for efficiency gains, unlike mice. To address this, the authors proposed two novel mechanisms that enable RL agents to exhibit more naturalistic risk-avoidance behaviors, such as cautious path planning and predator avoidance patterns.
Recent advances in reinforcement learning (RL) have demonstrated impressive capabilities in complex decision-making tasks. This progress raises a natural question: how do these artificial systems compare to biological agents, which have been shaped by millions of years of evolution? To help answer this question, we undertake a comparative study of biological mice and RL agents in a predator-avoidance maze environment. Through this analysis, we identify a striking disparity: RL agents consistently demonstrate a lack of self-preservation instinct, readily risking ``death'' for marginal efficiency gains. These risk-taking strategies are in contrast to biological agents, which exhibit sophisticated risk-assessment and avoidance behaviors. Towards bridging this gap between the biological and artificial, we propose two novel mechanisms that encourage more naturalistic risk-avoidance behaviors in RL agents. Our approach leads to the emergence of naturalistic behaviors, including strategic environment assessment, cautious path planning, and predator avoidance patterns that closely mirror those observed in biological systems.