Incentivizing exploration in reinforcement learning with deep predictive models
2017-08-13 19:00
537 查看
Stadie, Bradly C., Sergey Levine, and Pieter Abbeel. "Incentivizing exploration in reinforcement learning with deep predictive models." arXiv preprint arXiv:1507.00814 (2015).
作者通过模拟(状态，动作)的不确定性，从而修改reward，帮助agent进行探索。作者说用了他们的方法不用进行随机探索。该方法比较通用，适用于多种RL模型，但是要训练auto-encoder，所以也稍微有点繁琐。
实用指数：3颗星
理论指数：1颗星
创新指数：4颗星
作者通过模拟(状态，动作)的不确定性，从而修改reward，帮助agent进行探索。作者说用了他们的方法不用进行随机探索。该方法比较通用，适用于多种RL模型，但是要训练auto-encoder，所以也稍微有点繁琐。
实用指数：3颗星
理论指数：1颗星
创新指数：4颗星
内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理 
相关文章推荐
- using learning rate schedules for deep learning models in python with keras
- 下一代机器学习-在浏览器中训练深度学习模型Next Generation Machine Learning - Training Deep Learning Models in a Browser
- 论文笔记之：Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning
- Continuous control with Deep Reinforcement Learning与DDPG（Deep Deterministic Policy Gradient）的理解
- 18 Issues in Current Deep Reinforcement Learning from ZhiHu
- (转) Playing FPS games with deep reinforcement learning
- Playing Atari with Deep Reinforcement Learning算法解读
- 解读continuous control with deep reinforcement learning（DDPG）
- Playing Atari with Deep Reinforcement Learning
- PR10.10:#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
- 论文笔记之：Active Object Localization with Deep Reinforcement Learning
- DeepMind one shot learning 论文批注 One-Shot Generalization in Deep Generative Models
- Playing FPS Games with Deep Reinforcement Learning
- NOTE:Deep Reinforcement Learning with a Natural Language Action Space
- Deep Q-Network，NIPS-2013：Playing Atari with Deep Reinforcement Learning
- learning to communicate with deep multi-agent reinforcement learning
- Image Completion with Deep Learning in TensorFlow【DCGAN，图像补全】
- Paper Reading 3:Continuous control with Deep Reinforcement Learning
- 论文笔记之：Playing Atari with Deep Reinforcement Learning
- Continuous control with deep reinforcement learning(DDPG，深度确定策略梯度)练习