Print Email Facebook Twitter Conflicting demonstrations in Inverse Reinforcement Learning Title Conflicting demonstrations in Inverse Reinforcement Learning Author Labbé, Rafael (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Cavalcante Siebert, L. (mentor) Caregnato Neto, A. (mentor) Weber, J.M. (graduation committee) Degree granting institution Delft University of Technology Programme Computer Science and Engineering Project CSE3000 Research Project Date 2023-06-29 Abstract This paper aims to investigate the effect of conflicting demonstrations on Inverse Reinforcement Learning (IRL). IRL is a method to understand the intent of an expert, by only feeding it demonstrations of that expert, which may be a promising approach for areas such as self driving vehicles, where there are a lot of demonstrations from experts. This paper aims to investigate the effect of conflicting demonstrations on IRL. Demonstrations may not always come from the same expert or the expert may prioritize different goals at times. For example, a driver may not always do grocery shopping at the same store or they may take a slightly different route on different occasions. The results showcase a negative effect from severely conflicting demonstrations on the ability of Max Entropy IRL to recover rewards, but do show some slightly optimistic results on more than two goals. Subject Inverse Reinforcement LearningConflicting dataMaximum Entropy To reference this document use: http://resolver.tudelft.nl/uuid:8a452b02-0237-4131-b47d-92244c9916b1 Part of collection Student theses Document type bachelor thesis Rights © 2023 Rafael Labbé Files PDF CSE3000_Final_Paper_5_.pdf 849.91 KB Close viewer /islandora/object/uuid:8a452b02-0237-4131-b47d-92244c9916b1/datastream/OBJ/view