Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918027387207680 |
|---|---|
| author | Vasan, Gautham Elsayed, Mohamed Azimi, Alireza He, Jiamin Shariar, Fahim Bellinger, Colin White, Martha Mahmood, A. Rupam |
| author_facet | Vasan, Gautham Elsayed, Mohamed Azimi, Alireza He, Jiamin Shariar, Fahim Bellinger, Colin White, Martha Mahmood, A. Rupam |
| contents | Modern deep policy gradient methods achieve effective performance on simulated robotic tasks, but they all require large replay buffers or expensive batch updates, or both, making them incompatible for real systems with resource-limited computers. We show that these methods fail catastrophically when limited to small replay buffers or during incremental learning, where updates only use the most recent sample without batch updates or a replay buffer. We propose a novel incremental deep policy gradient method -- Action Value Gradient (AVG) and a set of normalization and scaling techniques to address the challenges of instability in incremental learning. On robotic simulation benchmarks, we show that AVG is the only incremental method that learns effectively, often achieving final performance comparable to batch policy gradient methods. This advancement enabled us to show for the first time effective deep reinforcement learning with real robots using only incremental updates, employing a robotic manipulator and a mobile robot. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_15370 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers Vasan, Gautham Elsayed, Mohamed Azimi, Alireza He, Jiamin Shariar, Fahim Bellinger, Colin White, Martha Mahmood, A. Rupam Machine Learning Artificial Intelligence Robotics Systems and Control Modern deep policy gradient methods achieve effective performance on simulated robotic tasks, but they all require large replay buffers or expensive batch updates, or both, making them incompatible for real systems with resource-limited computers. We show that these methods fail catastrophically when limited to small replay buffers or during incremental learning, where updates only use the most recent sample without batch updates or a replay buffer. We propose a novel incremental deep policy gradient method -- Action Value Gradient (AVG) and a set of normalization and scaling techniques to address the challenges of instability in incremental learning. On robotic simulation benchmarks, we show that AVG is the only incremental method that learns effectively, often achieving final performance comparable to batch policy gradient methods. This advancement enabled us to show for the first time effective deep reinforcement learning with real robots using only incremental updates, employing a robotic manipulator and a mobile robot. |
| title | Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers |
| topic | Machine Learning Artificial Intelligence Robotics Systems and Control |
| url | https://arxiv.org/abs/2411.15370 |