Repairing Learned Controllers with Convex Optimization: A Case Study

Guidotti D, Leofante F, Castellini C, Tacchella A (2019)

Publication Type: Conference contribution

Publication year: 2019

Journal

Lecture Notes in Computer Science Springer Verlag

Publisher: Springer Verlag

Book Volume: 11494 LNCS

Pages Range: 364-373

Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Event location: Thessaloniki GR

ISBN: 9783030192112

DOI: 10.1007/978-3-030-19212-9_24

Abstract

Despite the increasing popularity of Machine Learning methods, their usage in safety-critical applications is sometimes limited by the impossibility of providing formal guarantees on their behaviour. In this work we focus on one such application, where Kernel Ridge Regression with Random Fourier Features is used to learn controllers for a prosthetic hand. Due to the non-linearity of the activation function used, these controllers sometimes fail in correctly identifying users’ intention. Under specific circumstances muscular activation levels may be misinterpreted by the method, resulting in the prosthetic hand not behaving as intended. To alleviate this problem, we propose a novel method to verify the presence of this kind of intent detection mismatch and to repair controllers leveraging off-the-shelf LP technology without using additional data. We demonstrate the feasibility of our approach using datasets gathered from human participants.

Authors with CRIS profile

Claudio Castellini

Involved external institutions

University of Genova / Università degli Studi di Genova IT Italy (IT) Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR) / German Aerospace Center DE Germany (DE)

How to cite

APA:

Guidotti, D., Leofante, F., Castellini, C., & Tacchella, A. (2019). Repairing Learned Controllers with Convex Optimization: A Case Study. In Louis-Martin Rousseau, Kostas Stergiou (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 364-373). Thessaloniki, GR: Springer Verlag.

MLA:

Guidotti, Dario, et al. "Repairing Learned Controllers with Convex Optimization: A Case Study." Proceedings of the 16th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2019, Thessaloniki Ed. Louis-Martin Rousseau, Kostas Stergiou, Springer Verlag, 2019. 364-373.

BibTeX: Download