Sparse Representation Learning with Modified q-VAE towards Minimal Realization of World Model

TL;DR

This paper improves q-VAE to automatically produce sparse, minimal latent spaces for world models, enabling efficient real-time robot control with reduced dimensionality and faster task completion.

Contribution

It introduces a modified q-VAE that promotes sparsity, allowing automatic identification of the minimal latent space dimension for effective world modeling.

Findings

Successfully identified a six-dimensional latent space for a reaching task.

Reduced reaching time by approximately 20% using the minimal world model.

Demonstrated the method's ability to collapse unnecessary dimensions in the latent space.

Abstract

Extraction of low-dimensional latent space from high-dimensional observation data is essential to construct a real-time robot controller with a world model on the extracted latent space. However, there is no established method for tuning the dimension size of the latent space automatically, suffering from finding the necessary and sufficient dimension size, i.e. the minimal realization of the world model. In this study, we analyze and improve Tsallis-based variational autoencoder (q-VAE), and reveal that, under an appropriate configuration, it always facilitates making the latent space sparse. Even if the dimension size of the pre-specified latent space is redundant compared to the minimal realization, this sparsification collapses unnecessary dimensions, allowing for easy removal of them. We experimentally verified the benefits of the sparsification by the proposed method that it can easily find the necessary and sufficient six dimensions for a reaching task with a mobile manipulator that requires a six-dimensional state space. Moreover, by planning with such a minimal-realization world model learned in the extracted dimensions, the proposed method was able to exert a more optimal action sequence in real-time, reducing the reaching accomplishment time by around 20 %. The attached video is uploaded on youtube: https://youtu.be/-QjITrnxaRs