Inferring cost functions using reward parameter search and policy gradient reinforcement learning

Name: Inferring cost functions using reward parameter search and policy gradient reinforcement learning
Author: Arditi, Emir, Kunavar, T., Ugur, E., Babic, J., Öztop, Erhan

İsim	Inferring cost functions using reward parameter search and policy gradient reinforcement learning
Yazar	Arditi, Emir, Kunavar, T., Ugur, E., Babic, J., Öztop, Erhan
Basım Tarihi:	2021
Basım Yeri	- IEEE
Tür	Belge
Dil	İngilizce
Dijital	Evet
Yazma	Hayır
Kütüphane:	Özyeğin Üniversitesi
Demirbaş Numarası	978-1-6654-3554-3
Kayıt Numarası	23b429e6-82c9-4cfe-be04-500036388cd4
Lokasyon	Computer Science
Tarih	2021
Notlar	Slovenia/ARRS -Turkey/TUBITAK bilateral collaboration ; Bogazici Resarch Fund (BAP) IMAGINE-COG++ Project
Örnek Metin	This study focuses on inferring cost functions of obtained movement data using reward parameter search and policy gradient based Reinforcement Learning (RL). The behavior data for this task is obtained through a series of squat-to-stand movements of human participants under dynamic perturbations. The key parameter searched in the cost function is the weight of total torque used in performing the squat-to-stand action. An approximate model is used to learn squat-to-stand movements via a policy gradient method, namely Proximal Policy Optimization(PPO). A behavioral similarity metric based on Center of Mass(COM) is used to find the most likely weight parameter. The stochasticity in the training result of PPO is dealt with multiple runs, and as a result, a reasonable and a stable Inverse Reinforcement Learning(IRL) algorithm is obtained in terms of performance. The results indicate that for some participants, the reward function parameters of the experts were inferred successfully.
DOI	10.1109/IECON48115.2021.9589967

Kaynağa git Özyeğin Üniversitesi

Aramaya Dön

Özyeğin Üniversitesi

Kaynağa git

Inferring cost functions using reward parameter search and policy gradient reinforcement learning

Yazar Arditi, Emir, Kunavar, T., Ugur, E., Babic, J., Öztop, Erhan

Basım Tarihi 2021

Basım Yeri - IEEE

Tür Belge

Dil İngilizce

Dijital Evet

Yazma Hayır

Kütüphane Özyeğin Üniversitesi

Demirbaş Numarası 978-1-6654-3554-3

Kayıt Numarası 23b429e6-82c9-4cfe-be04-500036388cd4

Lokasyon Computer Science

Tarih 2021

Notlar Slovenia/ARRS -Turkey/TUBITAK bilateral collaboration ; Bogazici Resarch Fund (BAP) IMAGINE-COG++ Project

Örnek Metin This study focuses on inferring cost functions of obtained movement data using reward parameter search and policy gradient based Reinforcement Learning (RL). The behavior data for this task is obtained through a series of squat-to-stand movements of human participants under dynamic perturbations. The key parameter searched in the cost function is the weight of total torque used in performing the squat-to-stand action. An approximate model is used to learn squat-to-stand movements via a policy gradient method, namely Proximal Policy Optimization(PPO). A behavioral similarity metric based on Center of Mass(COM) is used to find the most likely weight parameter. The stochasticity in the training result of PPO is dealt with multiple runs, and as a result, a reasonable and a stable Inverse Reinforcement Learning(IRL) algorithm is obtained in terms of performance. The results indicate that for some participants, the reward function parameters of the experts were inferred successfully.

DOI 10.1109/IECON48115.2021.9589967