Instrumented data for causal scientific machine learning
Researchers have introduced a new concept called "instrumented data" for scientific machine learning, aiming to overcome limitations in current data types. This approach embeds the mechanistic model, its uncertainty, and counterfactuals directly with each data point. This method is expected to improve validation, auditing, and surrogate training across various scientific fields and could impact future foundation models for scientific reasoning. AI
IMPACT Introduces a novel data paradigm that could enhance the reliability and interpretability of AI models in scientific research.