Download presentation
Presentation is loading. Please wait.
Published byEmory Dalton Modified over 9 years ago
1
Human level control through deep reinforcement learning
Naiyan Wang
2
P 1 art Q Learning
3
Q Learning S A R tate ction eward
4
Q Learning Learning Rate Discount Factor New State Old State Reward
5
P 2 art Deep Q Learning
6
Traditional Cooking
7
Traditional Cooking
8
Traditional Cooking
9
Traditional Cooking
10
Traditional Cooking
11
End to End Cooking
12
End to End Learning
13
Formulation 1 2 3 Target Variable
14
Results Analysis DQN is good at … DQN is bad at …
15
P 3 art Discussion
16
Discussion Q: What is the key contributing factor?
A: Almost unlimited training data Q: How to account for long term dependency ? A: Long short term memory may be the solution
17
Thank You
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.