視線を用いた正則化による模倣学習の因果的混乱の緩和
GABRIL: Gaze-Based Regularization for Mitigating Causal Confusion in Imitation Learning
人間の視線データを模倣学習の正則化に利用し、因果的混乱を軽減する手法GABRILを提案。AtariとCARLAで有効性を検証。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Amin Banayeeanzade, Fatemeh Bahrani, Yutai Zhou, Erdem Bıyık
分類: cs.RO, cs.AI, cs.LG
原文アブストラクト
Imitation Learning (IL) is a widely adopted approach which enables agents to learn from human expert demonstrations by framing the task as a supervised learning problem. However, IL often suffers from causal confusion, where agents misinterpret spurious correlations as causal relationships, leading to poor performance in testing environments with distribution shift. To address this issue, we introduce GAze-Based Regularization in Imitation Learning (GABRIL), a novel method that leverages the human gaze data gathered during the data collection phase to guide the representation learning in IL. GABRIL utilizes a regularization loss which encourages the model to focus on causally relevant features identified through expert gaze and consequently mitigates the effects of confounding variables. We validate our approach in Atari environments and the Bench2Drive benchmark in CARLA by collecting human gaze datasets and applying our method in both domains. Experimental results show that the improvement of GABRIL over behavior cloning is around 179% more than the same number for other baselines in the Atari and 76% in the CARLA setup. Finally, we show that our method provides extra explainability when compared to regular IL agents.
