AI Research Associate for Early-Stage Scientific Discovery

TL;DR

This paper introduces an AI research associate designed for early-stage scientific discovery, utilizing a novel physics ontology, automated hypothesis generation, and interpretable tensor-based models to facilitate scientific insights from limited data.

Contribution

It presents a new physics-aware ontology, automated hypothesis search, and interpretable models for AI-driven scientific discovery, addressing limitations of existing AI methods.

Findings

Developed a context-aware physics ontology.

Automated generation of high-level hypotheses.

Constructed interpretable tensor-based models from sparse data.

Abstract

Artificial intelligence (AI) has been increasingly applied in scientific activities for decades; however, it is still far from an insightful and trustworthy collaborator in the scientific process. Most existing AI methods are either too simplistic to be useful in real problems faced by scientists or too domain-specialized (even dogmatized), stifling transformative discoveries or paradigm shifts. We present an AI research associate for early-stage scientific discovery based on (a) a novel minimally-biased ontology for physics-based modeling that is context-aware, interpretable, and generalizable across classical and relativistic physics; (b) automatic search for viable and parsimonious hypotheses, represented at a high-level (via domain-agnostic constructs) with built-in invariants, e.g., postulated forms of conservation principles implied by a presupposed spacetime topology; and (c) automatic compilation of the enumerated hypotheses to domain-specific, interpretable, and trainable/testable tensor-based computation graphs to learn phenomenological relations, e.g., constitutive or material laws, from sparse (and possibly noisy) data sets.