Difference between revisions of "Deep Q Network (DQN)"
| Line 7: | Line 7: | ||
When feedback is provided, it might be long time after the fateful decision has been made. In reality, the feedback is likely to be the result of a large number of prior decisions, taken amid a shifting, uncertain environment. Unlike supervised learning, there are no correct input/output pairs, so suboptimal actions are not explicitly corrected, wrong actions just decrease the corresponding value in the Q-table, meaning there’s less chance choosing the same action should the same state be encountered again. [http://www.quora.com/How-does-Q-learning-work-1 Quora | Jaron Collis] | When feedback is provided, it might be long time after the fateful decision has been made. In reality, the feedback is likely to be the result of a large number of prior decisions, taken amid a shifting, uncertain environment. Unlike supervised learning, there are no correct input/output pairs, so suboptimal actions are not explicitly corrected, wrong actions just decrease the corresponding value in the Q-table, meaning there’s less chance choosing the same action should the same state be encountered again. [http://www.quora.com/How-does-Q-learning-work-1 Quora | Jaron Collis] | ||
| + | |||
| + | Training deep neural networks to show that a novel end-to-end reinforcement learning agent, termed a deep Q-network (DQN) [http://deepmind.com/research/dqn/ Human-level control through Deep Reinforcement Learning | Deepmind] | ||
<youtube>79pmNdyxEGo</youtube> | <youtube>79pmNdyxEGo</youtube> | ||
Revision as of 05:42, 20 September 2018
Q Learning (DQN)
When feedback is provided, it might be long time after the fateful decision has been made. In reality, the feedback is likely to be the result of a large number of prior decisions, taken amid a shifting, uncertain environment. Unlike supervised learning, there are no correct input/output pairs, so suboptimal actions are not explicitly corrected, wrong actions just decrease the corresponding value in the Q-table, meaning there’s less chance choosing the same action should the same state be encountered again. Quora | Jaron Collis
Training deep neural networks to show that a novel end-to-end reinforcement learning agent, termed a deep Q-network (DQN) Human-level control through Deep Reinforcement Learning | Deepmind
