End-to-end reinforcement learning
 | This article, End-to-end reinforcement learning, has recently been created via the Articles for creation process. Please check to see if the reviewer has accidentally left this template after accepting the draft and take appropriate action as necessary.
Reviewer tools: Inform author |
In end-to-end reinfocement learning, the end-to-end process, in other words, the whole process from sensors(raw sensor signals or pixels) to motors (actions or motions) in a robot or agent is consisted of a layered or recurrent neural network, and is trained by reinforcement learning.[1] It has become popular through the learning of ATARI games[2][3] and Alpha-Go[4].
References
- ^ Demis Hassabis (2016). Artificial Intelligence and the Future. MIT Press. Online
- ^ V. Mnih et al. (2013). Playing atari with deep reinforcement learning. Online
- ^ V. Mnih et al. (2015). Human-level control through deep reinforcement learning. Nature 518, 529–533.[1]
- ^ D. Silver et al.(2016). Mastering the game of Go with deep neural networks and tree search. Nature 529, 484–489. [2]
| This article, End-to-end reinforcement learning, has recently been created via the Articles for creation process. Please check to see if the reviewer has accidentally left this template after accepting the draft and take appropriate action as necessary.
Reviewer tools: Inform author |