Summary of Model-based Reinforcement Learning For Parameterized Action Spaces, by Renhao Zhang et al.
Model-based Reinforcement Learning for Parameterized Action Spaces
by Renhao Zhang, Haotian Fu, Yilin Miao, George Konidaris
First submitted to arxiv on: 3 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel model-based reinforcement learning algorithm, Dynamics Learning and predictive control with Parameterized Actions (DLPA), is proposed for solving Parameterized Action Markov Decision Processes (PAMDPs). The DLPA agent learns a parameterized-action-conditioned dynamics model and plans using a modified Model Predictive Path Integral control. Theoretical analysis shows that the generated trajectory during planning is quantifiable in terms of Lipschitz Continuity. Empirically, our algorithm demonstrates superior sample efficiency and asymptotic performance compared to state-of-the-art PAMDP methods on various benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Our new algorithm helps robots and computers make better decisions by learning a model of how actions affect the world. This is useful for tasks like planning routes or controlling robots. The algorithm learns quickly and accurately, outperforming other similar approaches. |
Keywords
» Artificial intelligence » Reinforcement learning
