Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments

Kidziński Ł, Mohanty SP, Ong C, Huang Z, Zhou S, Pechenko A, Stelmaszczyk A, Jarosik P, Pavlov M, Kolesnikov S, Plis S, et al. (2018)
Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA.

Konferenzbeitrag | Veröffentlicht | Englisch
 
Autor*in
Kidziński, Łukasz; Mohanty, Sharada Prasanna; Ong, Carmichael; Huang, Zhewei; Zhou, Shuchang; Pechenko, Anton; Stelmaszczyk, Adam; Jarosik, Piotr; Pavlov, Mikhail; Kolesnikov, Sergey; Plis, Sergey; Chen, Zhibo
Alle
Abstract / Bemerkung
In the NIPS 2017 Learning to Run challenge, participants were tasked with building a controller for a musculoskeletal model to make it run as fast as possible through an obstacle course. Top participants were invited to describe their algorithms. In this work, we present eight solutions that used deep reinforcement learning approaches, based on algorithms such as Deep Deterministic Policy Gradient, Proximal Policy Optimization, and Trust Region Policy Optimization. Many solutions use similar relaxations and heuristics, such as reward shaping, frame skipping, discretization of the action space, symmetry, and policy blending. However, each of the eight teams implemented different modifications of the known algorithms.
Erscheinungsjahr
2018
Konferenz
31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track
Konferenzort
Long Beach, USA
Konferenzdatum
2017-12-04 – 2017-12-09
Page URI
https://pub.uni-bielefeld.de/record/2940661

Zitieren

Kidziński Ł, Mohanty SP, Ong C, et al. Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA.
Kidziński, Ł., Mohanty, S. P., Ong, C., Huang, Z., Zhou, S., Pechenko, A., Stelmaszczyk, A., et al. (2018). Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA.
Kidziński, Ł., Mohanty, S. P., Ong, C., Huang, Z., Zhou, S., Pechenko, A., Stelmaszczyk, A., Jarosik, P., Pavlov, M., Kolesnikov, S., et al. (2018).“Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments”. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA.
Kidziński, Ł., et al., 2018. Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA.
Ł. Kidziński, et al., “Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments”, Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA, 2018.
Kidziński, Ł., Mohanty, S.P., Ong, C., Huang, Z., Zhou, S., Pechenko, A., Stelmaszczyk, A., Jarosik, P., Pavlov, M., Kolesnikov, S., Plis, S., Chen, Z., Zhang, Z., Chen, J., Shi, J., Zheng, Z., Yuan, C., Lin, Z., Michalewski, H., Miłoś, P., Osiński, B., Melnik, A., Schilling, M., Ritter, H., Carroll, S., Hicks, J., Levine, S., Salathé, M., Delp, S.: Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA (2018).
Kidziński, Łukasz, Mohanty, Sharada Prasanna, Ong, Carmichael, Huang, Zhewei, Zhou, Shuchang, Pechenko, Anton, Stelmaszczyk, Adam, Jarosik, Piotr, Pavlov, Mikhail, Kolesnikov, Sergey, Plis, Sergey, Chen, Zhibo, Zhang, Zhizheng, Chen, Jiale, Shi, Jun, Zheng, Zhuobin, Yuan, Chun, Lin, Zhihui, Michalewski, Henryk, Miłoś, Piotr, Osiński, Błażej, Melnik, Andrew, Schilling, Malte, Ritter, Helge, Carroll, Sean, Hicks, Jennifer, Levine, Sergey, Salathé, Marcel, and Delp, Scott. “Learning to Run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments”. Presented at the 31nd Conference on Neural Information Processing Systems (NIPS 2017), Competition Track, Long Beach, USA, 2018.
Link(s) zu Volltext(en)
Access Level
OA Open Access

Export

Markieren/ Markierung löschen
Markierte Publikationen

Open Data PUB

Quellen

arXiv: 1804.00361

Suchen in

Google Scholar