Uncertainty Quantification in Scientific ML
Jayaraman J. Thiagarajan, USDOE National Nuclear Security Administration (NNSA), Rushil Anirudh, Peer‐Timo Bremer, Bindya Venkatesh · 2020
The intricate interactions between data sampling, model selection and the inherent randomness in complex systems strongly emphasize the need for a rigorous characterization of ML algorithms. In conventional statistics, uncertainty quantification (UQ) provides this characterization by measuring how accurately a model reflects the physical reality and by studying the impact of different error sources on the prediction.