Summary of Boosting Robustness in Preference-based Reinforcement Learning with Dynamic Sparsity, by Calarina Muslimani and Bram Grooten and Deepak Ranganatha Sastry Mamillapalli and Mykola Pechenizkiy and Decebal Constantin Mocanu and Matthew E. Taylor
Boosting Robustness in Preference-Based Reinforcement Learning with Dynamic Sparsity
by Calarina Muslimani, Bram Grooten, Deepak Ranganatha Sastry Mamillapalli, Mykola Pechenizkiy, Decebal Constantin Mocanu, Matthew E. Taylor
First submitted to arxiv on: 10 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 The paper proposes a new approach to preference-based reinforcement learning (PbRL) that enables autonomous agents to learn from humans in complex real-world environments. The authors focus on developing an algorithm, R2N (Robust-to-Noise), that can adapt to task-relevant features and ignore irrelevant distractions. This is achieved by leveraging dynamic sparse training principles to learn robust reward models. The authors demonstrate the effectiveness of R2N in a simulated teacher setting, outperforming state-of-the-art PbRL algorithms in multiple locomotion and control environments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps robots learn from people and adapt to their surroundings more effectively. It develops a new algorithm called R2N that can focus on what’s important and ignore distractions. This is useful for robots that need to work with people, like teaching robots or self-driving cars. |
Keywords
» Artificial intelligence » Reinforcement learning
