Operator Learning in Mechanics with Robust Error Estimation, Control, and Generalization

Organized minisymposium bringing together work on operator learning in mechanics where the emphasis is on reliability rather than raw accuracy: how to estimate the error a learned operator carries, how to control it, and what can be said about generalization beyond the distribution a surrogate was trained on.

Organized with Danial Faghihi and Serge Prudhomme. Two sessions on 23 June 2026, eight presentations, spanning Bayesian neural operators, residual-based error correction, reinforcement learning and robust generalization in mechanics.