On February 3, 2026, I had the opportunity to present our work at the Energy Consequences of Information Workshop 2026, held in Santa Fe, New Mexico, USA.
Our invited talk, “Energy-Efficient Solar Forecasting with a Neuromorphic–Bayesian State Estimation Approach,” explored how integrating reservoir computing with Bayesian filtering can improve forecasting accuracy while maintaining computational efficiency.
The core idea is to combine:
- Reservoir Computing (RC) for expressive and energy-efficient temporal modeling
- Bayesian state estimation for principled uncertainty quantification and robustness
By embedding neuromorphic-inspired dynamics within a probabilistic filtering framework, we can achieve improved predictive performance without sacrificing scalability — a key requirement for real-world energy systems and grid stability.
This work was conducted in collaboration with Prof. Francesco Sorrentino and Prof. Wenbin Wan (my PhD advisor) at the University of New Mexico.
Sincere thanks to the ECI organizers and sponsors — including the U.S. Department of Energy (DOE) Office of Science (ASCR), the Air Force Office of Scientific Research (AFOSR), the Air Force Research Laboratory (AFRL), and Sandia National Laboratories — for fostering such a thoughtful and interdisciplinary discussion at the energy–information nexus.
Slides and links Link to heading
- Slides (PDF): Download slides
- LinkedIn post: View on LinkedIn
If you attended the workshop and want to discuss the work, feel free to reach [email protected].