Pith. sign in

REVIEW 3 cited by

AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularity

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2006.10782 v2 pith:ZA6V5WW3 submitted 2020-06-18 cs.LG cs.AIcs.ITmath.ITphysics.comp-phstat.ML

Silviu-Marian Udrescu , Andrew Tan , Jiahai Feng , Orisvaldo Neto , Tailin Wu , Max Tegmark This is my paper · ORCID
classification cs.LGcs.AIcs.ITmath.ITphysics.comp-phstat.ML
keywords methodregressionsymbolicdatadiscoveringformulasgraphmodularity
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present an improved method for symbolic regression that seeks to fit data to formulas that are Pareto-optimal, in the sense of having the best accuracy for a given complexity. It improves on the previous state-of-the-art by typically being orders of magnitude more robust toward noise and bad data, and also by discovering many formulas that stumped previous methods. We develop a method for discovering generalized symmetries (arbitrary modularity in the computational graph of a formula) from gradient properties of a neural network fit. We use normalizing flows to generalize our symbolic regression method to probability distributions from which we only have samples, and employ statistical hypothesis testing to accelerate robust brute-force search.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A First Observational Assessment of Cosmic Backreaction Over an Extended Redshift Range

    astro-ph.CO 2026-04 unverdicted novelty 7.0 of 10

    First model-light constraints on Ω_R + 3Ω_Q from Pantheon+ and BAO reconstructions are consistent with flat FLRW yet too broad to exclude percent-level backreaction.

  2. A "Neural" Riemann solver for Relativistic Hydrodynamics

    gr-qc 2025-05 conditional novelty 6.0 of 10

    Small neural networks trained on exact solutions can replace the iterative root-finding in a relativistic Riemann solver, giving exact-like accuracy about 14 times faster in 1D tests.

  3. Bridging the Domain Gap in Equation Distillation with Reinforcement Feedback

    cs.LG 2025-05 conditional novelty 6.0 of 10

    Reinforcement learning fine-tuning with numerical fitness rewards improves equation discovery accuracy and noise robustness of a pretrained symbolic regression transformer.

Pith tools