深層方策勾配法の実態を徹底検証
A Closer Look at Deep Policy Gradients
深層方策勾配アルゴリズムの挙動を、勾配推定・価値予測・最適化景観の観点から詳細に分析し、理論的枠組みと実際の挙動が大きく乖離していることを示した研究。
著者: Andrew Ilyas, Logan Engstrom, Shibani Santurkar, Dimitris Tsipras, Firdaus Janoos, Larry Rudolph, Aleksander Madry
分類: cs.LG, cs.NE, cs.RO, stat.ML
原文アブストラクト
We study how the behavior of deep policy gradient algorithms reflects the conceptual framework motivating their development. To this end, we propose a fine-grained analysis of state-of-the-art methods based on key elements of this framework: gradient estimation, value prediction, and optimization landscapes. Our results show that the behavior of deep policy gradient algorithms often deviates from what their motivating framework would predict: the surrogate objective does not match the true reward landscape, learned value estimators fail to fit the true value function, and gradient estimates poorly correlate with the "true" gradient. The mismatch between predicted and empirical behavior we uncover highlights our poor understanding of current methods, and indicates the need to move beyond current benchmark-centric evaluation methods.
