Summary of Online Continual Learning For Interactive Instruction Following Agents, by Byeonghwi Kim et al.
Online Continual Learning For Interactive Instruction Following Agents
by Byeonghwi Kim, Minhyuk Seo, Jonghyun Choi
First submitted to arxiv on: 12 Mar 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG); Robotics (cs.RO)
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 In this paper, researchers explore a more realistic scenario for learning embodied agents that execute daily tasks via language directives. Instead of assuming the agent learns all training data at once, they propose two continual learning setups: Behavior Incremental Learning (Behavior-IL) and Environment Incremental Learning (Environment-IL). These methods aim to address limitations in previous “data prior” based continual learning approaches, which often rely on task boundary information that might not be available. To overcome this limitation, the authors introduce a new method called Confidence-Aware Moving Average (CAMA), which updates stored information using confidence scores during training. Empirical validations show that CAMA outperforms state-of-the-art methods by noticeable margins in both Behavior-IL and Environment-IL setups. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research focuses on creating more realistic embodied agents that can learn daily tasks through language directives. Normally, these agents learn all the information at once, but this isn’t how humans or robots really work. Instead, they propose two new ways for the agent to learn: Behavior Incremental Learning and Environment Incremental Learning. These methods are designed to be more flexible and adaptable than previous approaches. The researchers also introduce a new method called Confidence-Aware Moving Average, which helps the agent remember what it has learned without needing specific boundaries between tasks. |
Keywords
* Artificial intelligence * Continual learning
