Summary of Candere-coach: Reinforcement Learning From Noisy Feedback, by Yuxuan Li et al.
CANDERE-COACH: Reinforcement Learning from Noisy Feedback
by Yuxuan Li, Srijita Das, Matthew E. Taylor
First submitted to arxiv on: 23 Sep 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 The proposed CANDERE-COACH algorithm enables reinforcement learning (RL) agents to learn from noisy feedback provided by nonoptimal teachers, a common scenario in many challenging tasks. Building upon imitation learning, learning from preference, and inverse reinforcement learning frameworks, this method introduces a noise-filtering mechanism to de-noise online feedback data, allowing RL agents to successfully learn even with up to 40% of the teacher feedback being incorrect. This approach demonstrates effectiveness across three common domains. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers created a new way for computers to learn from imperfect teachers. Usually, computers need perfect feedback to learn well, but in real life, teachers can make mistakes or not be very good at giving feedback. The team developed an algorithm that helps the computer ignore bad feedback and still learn well. This works even when the teacher is wrong 40% of the time! They tested it on three different problems and found it worked great. |
Keywords
» Artificial intelligence » Reinforcement learning
