Summary of Behavior Generation with Latent Actions, by Seungjae Lee and Yibin Wang and Haritheja Etukuru and H. Jin Kim and Nur Muhammad Mahi Shafiullah and Lerrel Pinto
Behavior Generation with Latent Actions
by Seungjae Lee, Yibin Wang, Haritheja Etukuru, H. Jin Kim, Nur Muhammad Mahi Shafiullah, Lerrel Pinto
First submitted to arxiv on: 5 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 This paper presents Vector-Quantized Behavior Transformer (VQ-BeT), a novel generative model for decision-making that handles complex behaviors. Unlike language or image generation, decision-making requires modeling continuous-valued vectors with multimodal distributions. The proposed VQ-BeT model improves upon the existing Behavior Transformers (BeT) by tokenizing actions using hierarchical vector quantization, enabling it to capture behavior modes while accelerating inference speed 5x over Diffusion Policies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary VQ-BeT is a generative modeling approach for complex behaviors in decision-making. It addresses the limitations of previous models like BeT and Diffusion Policies by introducing a hierarchical vector quantization module that tokenizes continuous actions. This allows VQ-BeT to handle multimodal action prediction, conditional generation, and partial observations more effectively. |
Keywords
* Artificial intelligence * Diffusion * Generative model * Image generation * Inference * Quantization * Transformer
