Abstract
Scientific discovery systems typically optimize experiments within a fixed hypothesis space. This creates a failure mode when all available candidates omit the same missing mechanism: candidate disagreement can collapse even while the model class is systematically wrong. We formulate experimental model-class revision, in which a discovery policy jointly proposes a structural edit and a diagnostic experiment that tests whether that edit is necessary. The method couples a class-level distinguishability objective, in which one shared parameterization must explain all selected experiments, with anytime-valid sequential evidence that triggers structural revision only after the current class is rejected. On 400 held-out controlled dynamical environments, the joint policy reaches 89.5% exact recovery with a budget of 32 real experiments, improving the strongest matched baseline by 10.0 percentage points while requiring fewer executed experiments and candidate fits. The learned revision-experiment pairing transfers across unseen mechanism combinations, held-out but expressible primitives, parameter extrapolation, and shifted experiment costs; when the true mechanism is outside the edit grammar, it detects library insufficiency in 88% of cases with a 5.5% false-support rate. Revision gains also transfer to ODEBench and ODEBase model-library tasks, as well as DiscoverPhysics worlds. These results support a view of scientific discovery in which deciding what mechanisms a theory should make expressible and where to collect evidence are treated as a single sequential decision problem.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Ma, S., Gao, A., Zhu, C., Yan, X., Zhang, W., Zhang, W., Gong, L., Li, T., & Zhang, Q. (2026). Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space. https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space
MLA 9
Ma, Siyuan, et al. "Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space." https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space.
Chicago (author–date)
Ma, Siyuan, Albert Gao, Chunzheng Zhu, Xin Yan, Wenlong Zhang, Wenxin Zhang, Luqi Gong, Tianlin Li, and Qixin Zhang. 2026. "Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space." https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space.
Harvard
Ma, S., Gao, A., Zhu, C., Yan, X., Zhang, W., Zhang, W., Gong, L., Li, T. and Zhang, Q. (2026) 'Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space', Available at: https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space.
Vancouver
Ma S, Gao A, Zhu C, Yan X, Zhang W, Zhang W, et al. Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space. https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space
IEEE
S. Ma, A. Gao, C. Zhu, X. Yan, W. Zhang, W. Zhang, L. Gong, T. Li, and Q. Zhang, "Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space," https://omanscience.com/en/articles/learning-to-outgrow-a-theory-experimental-discovery-beyond-the-initial-hypothesis-space.