Machine Self-confidence in Autonomous Systems via Meta-analysis of Decision Processes
Algorithmic assurances assist human users in trusting advanced autonomous systems appropriately. This work explores one approach to creating assurances in which systems self-assess their decision-making capabilities, resulting in a ‘self-confidence’ measure. We present a framework for self-confidence assessment and reporting using meta-analysis factors, and then develop a new factor pertaining to ‘solver quality’ in the context of solving Markov decision processes (MDPs), which are widely used in autonomous systems. A novel method for computing solver quality self-confidence is derived, drawing inspiration from empirical hardness models. Numerical examples show our approach has desirable properties for enabling an MDP-based agent to self-assess its performance for a given task under different conditions. Experimental results for a simulated autonomous vehicle navigation problem show significantly improved delegated task performance outcomes in conditions where self-confidence reports are provided to users.
KeywordsHuman-Machine systems Artificial intelligence Self-assessment
- 3.Sweet, N., et al.: Towards self-confidence in autonomous systems. In: AIAA Infotech @ Aerospace, p. 1651 (2016)Google Scholar
- 6.Humphrey, L.: Model checking UAV mission plans. In: AIAA Modeling and Simulation Technologies Conference. Guidance, Navigation, and Control and Co-located Conferences. American Institute of Aeronautics and Astronautics, August 2012 Google Scholar
- 7.Aitken, M.: Assured human-autonomy interaction through machine self-confidence. M.S. thesis. University of Colorado at Boulder (2016)Google Scholar
- 8.Hutchins, A.R., et al.: Representing autonomous systems’ self-confidence through competency boundaries. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting, vol. 59, pp. 279–283. Sage (2015)Google Scholar
- 10.Israelsen, B.W., et al.: Factorized machine self-confidence for decision-making agents. arXiv:1810.06519 [cs.LG], October 2018