Summary of Latent Plan Transformer For Trajectory Abstraction: Planning As Latent Space Inference, by Deqian Kong et al.
Latent Plan Transformer for Trajectory Abstraction: Planning as Latent Space Inference
by Deqian Kong, Dehong Xu, Minglu Zhao, Bo Pang, Jianwen Xie, Andrew Lizarraga, Yuhao Huang, Sirui Xie, Ying Nian Wu
First submitted to arxiv on: 7 Feb 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 This research paper proposes the Latent Plan Transformer (LPT), a novel generative modeling approach for planning in long-term tasks. The LPT model leverages a latent variable to connect a trajectory generator and the final return, allowing it to be learned with maximum likelihood estimation on trajectory-return pairs. The authors demonstrate that the LPT can discover improved decisions from sub-optimal trajectories, achieving competitive performance across several benchmarks. This approach exhibits capabilities in nuanced credit assignments, trajectory stitching, and adaptation to environmental contingencies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study is about using computers to make better decisions by planning ahead. It introduces a new way to learn from data that helps robots or machines figure out what to do next based on past experiences. The method uses something called “latent variables” to connect what’s happened so far with what might happen in the future. This allows it to make smarter choices and adapt to changing situations. |
Keywords
* Artificial intelligence * Likelihood * Transformer
