状態依存重要度サンプリングによるブラックボックス自律システムの故障確率推定
Failure Probability Estimation for Black-Box Autonomous Systems using State-Dependent Importance Sampling Proposals
ブラックボックスな逐次意思決定システムの故障確率を、状態依存の適応的重要度サンプリングで高精度に推定する手法を提案した。
著者: Harrison Delecki, Sydney M. Katz, Mykel J. Kochenderfer
分類: cs.RO, cs.AI, cs.LG, stat.ML
原文アブストラクト
Estimating the probability of failure is a critical step in developing safety-critical autonomous systems. Direct estimation methods such as Monte Carlo sampling are often impractical due to the rarity of failures in these systems. Existing importance sampling approaches do not scale to sequential decision-making systems with large state spaces and long horizons. We propose an adaptive importance sampling algorithm to address these limitations. Our method minimizes the forward Kullback-Leibler divergence between a state-dependent proposal distribution and a relaxed form of the optimal importance sampling distribution. Our method uses Markov score ascent methods to estimate this objective. We evaluate our approach on four sequential systems and show that it provides more accurate failure probability estimates than baseline Monte Carlo and importance sampling techniques. This work is open sourced.
