エージェントの価値関数に潜む世界モデルの可視化
Visualizing the World Model in an Agent's Value Function
強化学習エージェントの価値関数が環境のルールや概念を内部に獲得しているかを、反事実的介入によって可視化・検証した研究。
著者: Shawheen Naderi, Fatemeh Akherati, Yubei Chen
分類: cs.AI, cs.LG
原文アブストラクト
Could model-free reinforcement learning algorithms actually be learning a model of the environment? In this work, we apply counterfactual interventions which hold everything about a state fixed, besides a targeted change, to explore if the value function of a Proximal Policy Optimization agent forms a world model of the Crafter environment. Despite only being trained to optimize reward, we discover that the value function learns concepts such as the rules governing when a diamond is collectible, water and food having an impact on health, heuristics regarding inventory, and more. Finally, we discuss the safety implications of our visualizations and some shortcomings of current reinforcement learning agents that will prevent us from building truly advanced machine intelligence.