"Operator Learning for Predictive Scientific Computing"
Lianghao Cao
California Institute of Technology
Wednesday, Jan 21, 2026
- Colloquium - 499 DSL Seminar Room
- 03:30 to 04:30 PM Eastern Time (US and Canada)
Click Here to Join via Zoom
Meeting # 942 7359 5552
Zoom Meeting # 942 7359 5552
Abstract:
Simulation-based predictions are increasingly driving high-stakes decisions that affect our welfare and security. Yet, building trust in these predictions remains a fundamental challenge. In this talk, I will discuss how operator learning can be tailored to enable reliable, predictive scientific computing that would otherwise be computationally intractable. I focus on the design and analysis of architectures and learning formulations for two distinct problems: inelastic homogenization in multiscale modeling, and amortized Bayesian inversion in uncertainty quantification. Finally, I outline my vision for integrating theory, experimentation, and computation to facilitate risk-aware, trustworthy decision-making.
| File | Description | File size |
|---|---|---|
 lianghaocao_profile.JPG | Headshot | 87 kB |
