Deep Learning in Geotechnical Engineering: A Critical Assessment of PINNs and Operator Learning
This work highlights the limitations of deep learning for geotechnical applications, cautioning against replacing proven solvers with slower, less accurate methods.
The paper critically assessed deep learning methods like PINNs and DeepONet in geotechnical engineering, finding they were up to 90,000 times slower and less accurate than traditional solvers, and recommended using automatic differentiation for inverse problems instead.
Deep learning methods -- physics-informed neural networks (PINNs), deep operator networks (DeepONet), and graph network simulators (GNS) -- are increasingly proposed for geotechnical problems. This paper tests these methods against traditional solvers on canonical problems: wave propagation and beam-foundation interaction. PINNs run 90,000 times slower than finite difference with larger errors. DeepONet requires thousands of training simulations and breaks even only after millions of evaluations. Multi-layer perceptrons fail catastrophically when extrapolating beyond training data -- the common case in geotechnical prediction. GNS shows promise for geometry-agnostic simulation but faces scaling limits and cannot capture path-dependent soil behavior. For inverse problems, automatic differentiation through traditional solvers recovers material parameters with sub-percent accuracy in seconds. We recommend: use automatic differentiation for inverse problems; apply site-based cross-validation to account for spatial autocorrelation; reserve neural networks for problems where traditional solvers are genuinely expensive and predictions remain within the training envelope. When a method is four orders of magnitude slower with less accuracy, it is not a viable replacement for proven solvers.