Meta-Learning for Physically-Constrained Neural System Identification
This work addresses the challenge of learning from limited target data in system identification, which is incremental as it builds on existing meta-learning and neural modeling approaches.
The paper tackles the problem of rapid adaptation of neural state-space models for black-box system identification by using a gradient-based meta-learning framework that leverages data from diverse source systems, resulting in improved downstream performance in model-based state estimation for indoor localization and energy systems.
We present a gradient-based meta-learning framework for rapid adaptation of neural state-space models (NSSMs) for black-box system identification. When applicable, we also incorporate domain-specific physical constraints to improve the accuracy of the NSSM. The major benefit of our approach is that instead of relying solely on data from a single target system, our framework utilizes data from a diverse set of source systems, enabling learning from limited target data, as well as with few online training iterations. Through benchmark examples, we demonstrate the potential of our approach, study the effect of fine-tuning subnetworks rather than full fine-tuning, and report real-world case studies to illustrate the practical application and generalizability of the approach to practical problems with physical-constraints. Specifically, we show that the meta-learned models result in improved downstream performance in model-based state estimation in indoor localization and energy systems.