Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

VQ-CNMP: Neuro-Symbolic Skill Learning for Bi-Level Planning

Hakan Aktas, Emre Ugur

TL;DR

A novel neural network model capable of discovering high-level skill representations from unlabeled demonstration data is proposed and a bi-level planning pipeline that utilizes the model using a gradient-based planning approach is proposed.

Abstract

This paper proposes a novel neural network model capable of discovering high-level skill representations from unlabeled demonstration data. We also propose a bi-level planning pipeline that utilizes our model using a gradient-based planning approach. While extracting high-level representations, our model also preserves the low-level information, which can be used for low-level action planning. In the experiments, we tested the skill discovery performance of our model under different conditions, tested whether Multi-Modal LLMs can be utilized to label the learned high-level skill representations, and finally tested the high-level and low-level planning performance of our pipeline.

VQ-CNMP: Neuro-Symbolic Skill Learning for Bi-Level Planning

TL;DR

A novel neural network model capable of discovering high-level skill representations from unlabeled demonstration data is proposed and a bi-level planning pipeline that utilizes the model using a gradient-based planning approach is proposed.

Abstract

This paper proposes a novel neural network model capable of discovering high-level skill representations from unlabeled demonstration data. We also propose a bi-level planning pipeline that utilizes our model using a gradient-based planning approach. While extracting high-level representations, our model also preserves the low-level information, which can be used for low-level action planning. In the experiments, we tested the skill discovery performance of our model under different conditions, tested whether Multi-Modal LLMs can be utilized to label the learned high-level skill representations, and finally tested the high-level and low-level planning performance of our pipeline.

Paper Structure

This paper contains 20 sections, 6 equations, 6 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Model
  4. Self-Supervised Fine-Tuning
  5. Planning
  6. Experimental Results
  7. Skill Discovery
  8. Labeling using Multi-Modal LLMs
  9. Planning Performance
  10. High-Level Planning Performance
  11. Effect of Number of Skills in the Skill Space
  12. Environment with Hidden Elements
  13. Low-Level Planning Performance
  14. Conclusion, Limitations, Future Work and Discussion
  15. Experimental Results
  16. ...and 5 more sections

Figures (6)

  • Figure 1: An overview of a generalist agent that uses abstractions to interact with its environment.
  • Figure 2: The Overview of our Model
  • Figure 3: Planning system
  • Figure 4: Adding oil and potato to the pan actions.
  • Figure 5: Image of the environment used for high-level planning.
  • ...and 1 more figures