We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Robotics

Title: The Adaptive Stress Testing Formulation

Abstract: Validation is a key challenge in the search for safe autonomy. Simulations are often either too simple to provide robust validation, or too complex to tractably compute. Therefore, approximate validation methods are needed to tractably find failures without unsafe simplifications. This paper presents the theory behind one such black-box approach: adaptive stress testing (AST). We also provide three examples of validation problems formulated to work with AST.
Comments: Presented at the Workshop on Robust Autonomy at RSS 2019
Subjects: Robotics (cs.RO); Machine Learning (cs.LG); Systems and Control (eess.SY); Machine Learning (stat.ML)
Cite as: arXiv:2004.04293 [cs.RO]
  (or arXiv:2004.04293v1 [cs.RO] for this version)

Submission history

From: Mark Koren [view email]
[v1] Wed, 8 Apr 2020 23:04:42 GMT (203kb,D)

Link back to: arXiv, form interface, contact.