Meta-learning using privileged information for dynamics

TL;DR

This paper introduces an extension of Neural ODE Processes that incorporates privileged information, such as structured knowledge, to improve meta-learning of dynamics, demonstrating enhanced accuracy and calibration in simulated tasks.

Contribution

The paper proposes a novel method to integrate privileged information into Neural ODE Processes for better dynamics modeling in meta-learning.

Findings

Improved accuracy on simulated dynamics tasks

Enhanced calibration of the model predictions

Effective use of structured knowledge as privileged information

Abstract

Neural ODE Processes approach the problem of meta-learning for dynamics using a latent variable model, which permits a flexible aggregation of contextual information. This flexibility is inherited from the Neural Process framework and allows the model to aggregate sets of context observations of arbitrary size into a fixed-length representation. In the physical sciences, we often have access to structured knowledge in addition to raw observations of a system, such as the value of a conserved quantity or a description of an understood component. Taking advantage of the aggregation flexibility, we extend the Neural ODE Process model to use additional information within the Learning Using Privileged Information setting, and we validate our extension with experiments showing improved accuracy and calibration on simulated dynamics tasks.