Mostra el registre d'ítem simple
Evaluating the Robustness of GAN-Based Inverse Reinforcement Learning Algorithms
dc.contributor | Martín Muñoz, Mario |
dc.contributor | Garcia Gasulla, Dario |
dc.contributor.author | Heidecke, Johannes |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament de Ciències de la Computació |
dc.date.accessioned | 2019-04-11T09:17:00Z |
dc.date.available | 2019-04-11T09:17:00Z |
dc.date.issued | 2019-01-15 |
dc.identifier.uri | http://hdl.handle.net/2117/131625 |
dc.description.abstract | We evaluate the robustness of reward functions learned with IRL, when transferred to similar tasks. We exceed state of the art results for one benchmark task and solve another one for the first time. Modifications are proposed that achieve faster and more stable training. |
dc.language.iso | eng |
dc.publisher | Universitat Politècnica de Catalunya |
dc.subject | Àrees temàtiques de la UPC::Informàtica |
dc.subject.lcsh | Reinforcement learning |
dc.subject.lcsh | Algorithms |
dc.subject.other | inverse reinforcement learning |
dc.subject.other | IRL |
dc.subject.other | reinforcement learning |
dc.subject.other | RL |
dc.subject.other | guided cost learning |
dc.subject.other | GCL |
dc.subject.other | adversarial inverse reinforcement learning |
dc.subject.other | AIRL |
dc.subject.other | soft actor critic |
dc.subject.other | SAC |
dc.subject.other | transfer learning |
dc.subject.other | robustness |
dc.subject.other | maximum entropy principle |
dc.subject.other | maximum causal entropy principle |
dc.subject.other | reward shaping |
dc.subject.other | pre-training |
dc.subject.other | metric |
dc.subject.other | shaped reward loss |
dc.subject.other | pendulum |
dc.subject.other | lunar lander |
dc.title | Evaluating the Robustness of GAN-Based Inverse Reinforcement Learning Algorithms |
dc.type | Master thesis |
dc.subject.lemac | Aprenentatge per reforç |
dc.subject.lemac | Algorismes |
dc.identifier.slug | 134533 |
dc.rights.access | Open Access |
dc.date.updated | 2019-02-04T05:00:43Z |
dc.audience.educationlevel | Màster |
dc.audience.mediator | Facultat d'Informàtica de Barcelona |
dc.audience.degree | MÀSTER UNIVERSITARI EN INTEL·LIGÈNCIA ARTIFICIAL (Pla 2017) |
dc.contributor.covenantee | Universitat de Barcelona |
dc.contributor.covenantee | Universitat Rovira i Virgili |