Deep Reinforcement Learning
Deep reinforcement learning is Reinforcement Learning using deep Neural Networks. It is a field popularized by (Mnih et al. 2015), but using Connectionist Networks in RL has been persued much earlier (Tesauro 1994), (Bakker 2002), (Rummery and Niranjan 1994), (Williams 1992), (Liu and Zou 2017) and many others.
References
Bakker, Bram. 2002. “Reinforcement Learning with Long Short-Term Memory.” In Advances in Neural Information Processing Systems 14, 1475–82. MIT Press.
Mnih, Volodymyr, Koray Kavukcuoglu, David Silver, Andrei A. Rusu, Joel Veness, Marc G. Bellemare, Alex Graves, et al. 2015. “Human-Level Control through Deep Reinforcement Learning.” Nature 518 (7540): 529–33. doi:10.1038/nature14236.
Rummery, Gavin A, and Mahesan Niranjan. 1994. On-Line Q-learning Using Connectionist Systems. Vol. 37. University of Cambridge, Department of Engineering Cambridge, UK.
Tesauro, Gerald. 1994. “TD-Gammon, a Self-Teaching Backgammon Program, Achieves Master-Level Play.” Neural Computation 6: 215–19.
Williams, Ronald J. 1992. “Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning.” In Reinforcement Learning, 5–32. The Springer International Series in Engineering and Computer Science. Springer, Boston, MA. doi:10.1007/978-1-4615-3618-5_2.
Links to this note:
- Reinforcement Learning
- veeriah2019discovery: Discovery of Useful Questions as Auxiliary Tasks
- mnih2016asynchronous: Asynchronous Methods for Deep Reinforcement Learning
- machado2018revisiting: Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents
- jaderberg2017reinforcement: Reinforcement Learning with Unsupervised Auxiliary Tasks
- henderson2018deep: Deep Reinforcement Learning That Matters
- Experience Replay
- DeepMind Lab
- Auxiliary Tasks
- Atari