Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2407.13755 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912248980570112 |
|---|---|
| author | Mahankali, Srinath Hong, Zhang-Wei Sekhari, Ayush Rakhlin, Alexander Agrawal, Pulkit |
| author_facet | Mahankali, Srinath Hong, Zhang-Wei Sekhari, Ayush Rakhlin, Alexander Agrawal, Pulkit |
| contents | We introduce Random Latent Exploration (RLE), a simple yet effective exploration strategy in reinforcement learning (RL). On average, RLE outperforms noise-based methods, which perturb the agent's actions, and bonus-based exploration, which rewards the agent for attempting novel behaviors. The core idea of RLE is to encourage the agent to explore different parts of the environment by pursuing randomly sampled goals in a latent space. RLE is as simple as noise-based methods, as it avoids complex bonus calculations but retains the deep exploration benefits of bonus-based methods. Our experiments show that RLE improves performance on average in both discrete (e.g., Atari) and continuous control tasks (e.g., Isaac Gym), enhancing exploration while remaining a simple and general plug-in for existing RL algorithms. Project website and code: https://srinathm1359.github.io/random-latent-exploration |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_13755 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Random Latent Exploration for Deep Reinforcement Learning Mahankali, Srinath Hong, Zhang-Wei Sekhari, Ayush Rakhlin, Alexander Agrawal, Pulkit Machine Learning We introduce Random Latent Exploration (RLE), a simple yet effective exploration strategy in reinforcement learning (RL). On average, RLE outperforms noise-based methods, which perturb the agent's actions, and bonus-based exploration, which rewards the agent for attempting novel behaviors. The core idea of RLE is to encourage the agent to explore different parts of the environment by pursuing randomly sampled goals in a latent space. RLE is as simple as noise-based methods, as it avoids complex bonus calculations but retains the deep exploration benefits of bonus-based methods. Our experiments show that RLE improves performance on average in both discrete (e.g., Atari) and continuous control tasks (e.g., Isaac Gym), enhancing exploration while remaining a simple and general plug-in for existing RL algorithms. Project website and code: https://srinathm1359.github.io/random-latent-exploration |
| title | Random Latent Exploration for Deep Reinforcement Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2407.13755 |