Improve this page

ML Track

The Machine Learning Track is to award the team that conducts and reports the most novel machine learning study using our simulation environment. To win this track, one should:

  • rank in the top 50 of Round 1
  • submit a paper describing your machine learning approach in NeurIPS format

Our review board led by Xue Bin (Jason) Peng, xbpeng@berkeley.edu) will select the paper that presents the most novel machine learning study using our simulation environment. The topics of interest include but are not limited to:

  • reinforcement learning (e.g. novel RL algorithms or improvements to existing techniques [Kakade, 2001; Silver et al., 2014; Haarnoja et al., 2018])
  • imitation learning (e.g. methods for learning from demonstrations [Peng et al., 2018; Ho and Ermon, 2016])
  • reward functions (e.g. designing reward function that encourage natural behaviours [Wang et al., 2009; Yu et al., 2018])
  • control structures (e.g. task-specific control structures for motion control [Wang et al., 2018; Lee et al., 2019])

The prizes for the winner are:

  • 2x NVIDIA GPU
  • Travel grant (up to $1,000 to NeurIPS 2019 or Stanford)
  • Paper acceptance into the NeurIPS Deep RL workshop

References

  • Haarnoja, T., Zhou, A., Abbeel, P., & Levine, S. (2018). Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor. Proceedings of the 35th International Conference on Machine Learning (ICML 2018).
  • Ho, J., & Ermon, S. (2016). Generative Adversarial Imitation Learning. Advances in Neural Information Processing Systems (NIPS 2016).
  • Kakade, S. (2001). A Natural Policy Gradient. Proceedings of the 14th International Conference on Neural Information Processing Systems (NIPS 2001).
  • Lee, S., Lee, K., Park, M., and Lee, J. (2019). Scalable Muscle-actuated Human Simulation and Control. ACM Transactions on Graphics (Proc. SIGGRAPH 2019).
  • Peng, X. B., Abbeel, P., Levine, S., & van de Panne, M. (2018). Deepmimic: Example-guided deep reinforcement learning of physics-based character skills. ACM Transactions on Graphics (Proc. SIGGRAPH 2018).
  • Silver, D., Lever, G., Heess, N., Degris, T., Weirstra, D., & Riedmiller, M. (2014). Deterministic Policy Gradient Algorithms. Proceedings of the 31st International Conference on Machine Learning (ICML 2014).
  • Yu, W., Turk, G., & Liu, C.K. (2018). Learning Symmetric and Low-Energy Locomotion. ACM Transactions on Graphics (Proc. SIGGRAPH 2018).
  • Wang, J., Fleet, D., & Hertzmann, A. (2009). Optimizing Walking Controllers. ACM Transactions on Graphics (Proc. SIGGRAPH Asia 2009).
  • Wang, T., Liao, R., Ba, J., & Fidler, S. (2018). NerveNet: Learning Structured Policy with Graph Neural Networks. International Conference on Learning Representations (ICLR 2019).