Using reward-weighted imitations for robot reinforcement learning

2009 IEEE Symposium on Adaptive Dynamic Programming and Reinforcement Learning, ADPRL 2009 - Proceedings

Volumn , Issue , 2009, Pages 226-232

  Peters, Jan ^a   Kober, Jens ^a  

^a MAX PLANCK INSTITUTE FOR BIOLOGICAL CYBERNETICS   (Germany)

Author keywords

[No Author keywords available]

Indexed keywords

LEARNING TASKS; MOTOR PRIMITIVES; MOTOR SKILLS; NOVEL ALGORITHM; REINFORCEMENT LEARNING APPROACH; ROBOT ARMS;

AUTOMOBILE EXHIBITIONS; DYNAMIC PROGRAMMING; EDUCATION; MOTORS; REINFORCEMENT; REINFORCEMENT LEARNING; ROBOTS; SYSTEMS ENGINEERING;

LEARNING ALGORITHMS;

EID: 67650458573     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ADPRL.2009.4927549     Document Type: Conference Paper

References (25)

1. R. Sutton and A. Bano. Reinforcement Learning. MIT Press. 1998.
2. J. BagnelL S. Kadade, A. Ng, and J. Sehneidcr. Policy sem-ch hy dynamic programming. In Advances in Neural Information Processing Systems (NIPS), 2003.
3. A. Ng and M. Jordan. PEGASUS: A policy sem-ch method for largc MDPs and pOMDPs. In International Conference on Uncertainty in Artificial Intelligence (UAl), 2000.
4. F. Guenter. M. Hersch. S. Cali non. and A. Billard. Reinforcement learning for imitating constrained reaching movements. RSf Advanced Robotics. 21. 1521-1544. 2007. (Pubitemid 47529845)
5. M. Hollman. A. Douect. N. de Freitas. and A. Jasra. Bayesian policy leal11ing with trans-dimensional MCMC. In Advances in Neural Information Processing Systems (Nfl'S). 2007.
6. R. I. Williams. Simple statistical gradient-following algorithms for conncctionist reinforcemcnt leal11ing. Machine Learning, 8 :229-256, 1992.
7. R. S. Sutton, D. McAllester, S. Singh, and Y. Mansour. Policy gradient methods for reinforcement learning with function approximation. In Advances in Neural Information I'rocessing Systems (NIl'S), 2000.
8. J. Bagnell and J. Schneider. Covariant policy search. In International .Joint Conference on Artificial Intelligence (J.JCAI), 2003.
9. J. Peters and S. Schaa Policy gradient methods for robotics. In International Conference on Intelligent Robots and Systems (IROS), 2006. (Pubitemid 46928224)
10. G. Wulf Attention and motor skill learning. Human Kinetics. Champaign. IL., 2007.
11. G.J. McLachan and T. Krishnan. The EM Algorithm and Extensions. Wiley Series in Prohahility and Statistics. John Wiley & Sons, 1997.
12. P. Dayan and G. E. Hinton. Using expectation-maximization for reinforcement learning. Neural Computation, 9(2):271-278, 1997. (Pubitemid 127635391)
13. T. Ruckstieb, M. Felder, and J. Schmidhuher. State-Dependent Exploration for Policy Gradient Methods. In Proceedings of the European Conference on Machine Learning (ECML), Antwerp, HelgiLlln, 2008.
14. J. Peters and S. Schaal. Learning operational space control. In Procccdings of Rohotics: Sciencc and Systcms (RSS), Philadelphia PA, 2006.
15. M. Kawato, F. Gandolfo, H. Gomi, and Y. Wada. Teaching by showing in kendama based on optimization principle. In International Conf on Anilicial Neural Networks, 1994.
16. C. G. Atkeson. Using local trajectory optimizer.I' to speed up global optimization in dynamic programming. In Advances in Neural Information Processing Systems (NIPS), 1994.
17. J. Kober, J. Peters. Policy Search of Motor Primitives for Robotics. In Advanccs in Neural Information proccssing Systcms (NIPS), 2008.
18. A. Ijspeert, J. Nakanishi, and S. Schaal. Learning attractor landscapes for learning motor primitives. In Advanccs in Ncural Inlormation Proccssing Systcms (NIPS), 2003.
19. S. Schaal, P. Mohajerian, and A. Ijspeert. Dynamics systems V.I'. optimal control - a unifying view. Progrcss in Brain Rcsem-ch, 165(1):425-445, 2007. (Pubitemid 47513886)
20. Wikipedia, May 31, 2008. http:I//en.wikipedia.org/wiki/Ball-in-aCUP
21. L Sciavicco and B. Siciliano, Modcling and control of rohot manipulators. Heidelberg, Germany: MacGraw-Hill, 2007.
22. A. D'Souza, S. Vijayakumw; and S. Schaa Learning inverse kinematics, " In Proccedings of the IEEE/RSJ International Confercnce on Intelligent Robots and Systems (IROS), Hawaii, USI, 2001.
23. D. DeMers and K. Kreutz-Delgado. Canonical parameterization of excess motor degrees offreedOln with self organizing maps. In IEEE Transactions on Neural Networks, Vol. 7, PP'. 43-55, 1996. (Pubitemid 126811727)
24. J. Nakanishi, M. Mistry, J. Peters, and S. Schaal. Operational space control: A theoretical and emprical comparison. In International Journal of Robotics Rcsearch (I.JRR), pp.737-757, Vol 27 (6), 2008.
25. J. Peters, M. Mistry', F. Udwadia, R. Cory, J. Nakanishi, and S. Schaal. A unifying methodology fin robot control with redundant DOFs In Autonomous Robots, pp.1-12, Vol 24 (1), 2008. (Pubitemid 350276040)