Model-based Validation as Probabilistic Inference

被引:0
|
作者
Delecki, Harrison [1 ]
Corso, Anthony [1 ]
Kochenderfer, Mykel J. [1 ]
机构
[1] Stanford Univ, Dept Aeronaut & Astronaut, Stanford, CA 94305 USA
基金
美国国家科学基金会;
关键词
safety validation; Bayesian inference;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Estimating the distribution over failures is a key step in validating autonomous systems. Existing approaches focus on finding failures for a small range of initial conditions or make restrictive assumptions about the properties of the system under test. We frame estimating the distribution over failure trajectories for sequential systems as Bayesian inference. Our model-based approach represents the distribution over failure trajectories using rollouts of system dynamics and computes trajectory gradients using automatic differentiation. Our approach is demonstrated in an inverted pendulum control system, an autonomous vehicle driving scenario, and a partially observable lunar lander. Sampling is performed using an off-the-shelf implementation of Hamiltonian Monte Carlo with multiple chains to capture multimodality and gradient smoothing for safe trajectories. In all experiments, we observed improvements in sample efficiency and parameter space coverage compared to black-box baseline approaches. This work is open sourced.
引用
收藏
页数:13
相关论文
共 50 条
  • [31] In defence of model-based inference in phylogeography REPLY
    Beaumont, Mark A.
    Nielsen, Rasmus
    Robert, Christian
    Hey, Jody
    Gaggiotti, Oscar
    Knowles, Lacey
    Estoup, Arnaud
    Panchal, Mahesh
    Corander, Jukka
    Hickerson, Mike
    Sisson, Scott A.
    Fagundes, Nelson
    Chikhi, Lounes
    Beerli, Peter
    Vitalis, Renaud
    Cornuet, Jean-Marie
    Huelsenbeck, John
    Foll, Matthieu
    Yang, Ziheng
    Rousset, Francois
    Balding, David
    Excoffier, Laurent
    MOLECULAR ECOLOGY, 2010, 19 (03) : 436 - 446
  • [32] Probabilistic Mixture Model-Based Spectral Unmixing
    Hoidn, Oliver
    Mishra, Aashwin Ananda
    Mehta, Apurva
    APPLIED SCIENCES-BASEL, 2024, 14 (11):
  • [33] A Probabilistic Framework for Model-Based Imitation Learning
    Shon, Aaron P.
    Grimes, David B.
    Baker, Chris L.
    Rao, Rajesh P. N.
    PROCEEDINGS OF THE TWENTY-SIXTH ANNUAL CONFERENCE OF THE COGNITIVE SCIENCE SOCIETY, 2004, : 1237 - 1242
  • [34] A model-based algorithm for the Probabilistic Orienteering Problem
    Montemanni, Roberto
    Smith, Derek H.
    COMPUTERS & OPERATIONS RESEARCH, 2025, 176
  • [35] Residual autocorrelation in probabilistic model-based diagnostics
    Schwall, Matthew L.
    Gerdes, J. Christian
    PROCEEDINGS OF THE ASME DYNAMIC SYSTEMS AND CONTROL DIVISION 2005, PTS A AND B, 2005, : 1255 - 1264
  • [36] Model-based probabilistic frequent itemset mining
    Thomas Bernecker
    Reynold Cheng
    David W. Cheung
    Hans-Peter Kriegel
    Sau Dan Lee
    Matthias Renz
    Florian Verhein
    Liang Wang
    Andreas Zuefle
    Knowledge and Information Systems, 2013, 37 : 181 - 217
  • [37] Model-based probabilistic frequent itemset mining
    Bernecker, Thomas
    Cheng, Reynold
    Cheung, David W.
    Kriegel, Hans-Peter
    Lee, Sau Dan
    Renz, Matthias
    Verhein, Florian
    Wang, Liang
    Zuefle, Andreas
    KNOWLEDGE AND INFORMATION SYSTEMS, 2013, 37 (01) : 181 - 217
  • [38] Model-based validation of a DOx sensor
    Clarke, DW
    Fraher, PMA
    CONTROL ENGINEERING PRACTICE, 1996, 4 (09) : 1313 - 1320
  • [39] Model-based validation of CANopen systems
    Lekidis, Alexios
    Bozga, Marius
    Bensalem, Saddek
    PROCEEDINGS OF 2014 10TH IEEE WORKSHOP ON FACTORY COMMUNICATION SYSTEMS (WFCS 2014), 2014,
  • [40] Model-Based Validation for Internet Services
    Tjang, Andrew
    Oliveira, Fabio
    Bianchini, Ricardo
    Martin, Richard P.
    Nguyen, Thu D.
    2009 28TH IEEE INTERNATIONAL SYMPOSIUM ON RELIABLE DISTRIBUTED SYSTEMS, PROCEEDINGS, 2009, : 61 - 70