Exploring Temporal Representation in Neural Processes for Multimodal Action Prediction

TL;DR

This paper investigates the use of Conditional Neural Processes for multimodal action prediction in robotics, proposing a temporal encoding enhancement to improve generalization to unseen actions.

Contribution

It introduces DMBN-PTE, a revised model with improved temporal representation, advancing autonomous action forecasting in robotic systems.

Findings

DMBN can reconstruct visuo-motor signals during partial actions.

DMBN struggles to generalize to unseen actions due to temporal representation issues.

DMBN-PTE improves temporal robustness and prediction capabilities.

Abstract

Inspired by the human ability to understand and predict others, we study the applicability of Conditional Neural Processes (CNP) to the task of self-supervised multimodal action prediction in robotics. Following recent results regarding the ontogeny of the Mirror Neuron System (MNS), we focus on the preliminary objective of self-actions prediction. We find a good MNS-inspired model in the existing Deep Modality Blending Network (DMBN), able to reconstruct the visuo-motor sensory signal during a partially observed action sequence by leveraging the probabilistic generation of CNP. After a qualitative and quantitative evaluation, we highlight its difficulties in generalizing to unseen action sequences, and identify the cause in its inner representation of time. Therefore, we propose a revised version, termed DMBN-Positional Time Encoding (DMBN-PTE), that facilitates learning a more robust representation of temporal information, and provide preliminary results of its effectiveness in expanding the applicability of the architecture. DMBN-PTE figures as a first step in the development of robotic systems that autonomously learn to forecast actions on longer time scales refining their predictions with incoming observations.