[2603.21836] Instruction Set and Language for Symbolic Regression

Computer Science > Computation and Language arXiv:2603.21836 (cs) [Submitted on 23 Mar 2026] Title:Instruction Set and Language for Symbolic Regression Authors:Ezequiel Lopez-Rubio, Mario Pascual-Gonzalez View a PDF of the paper titled Instruction Set and Language for Symbolic Regression, by Ezequiel Lopez-Rubio and 1 other authors View PDF HTML (experimental) Abstract:A fundamental but largely unaddressed obstacle in Symbolic regression (SR) is structural redundancy: every expression DAG with admits many distinct node-numbering schemes that all encode the same expression, each occupying a separate point in the search space and consuming fitness evaluations without adding diversity. We present IsalSR (Instruction Set and Language for Symbolic Regression), a representation framework that encodes expression DAGs as strings over a compact two-tier alphabet and computes a pruned canonical string -- a complete labeled-DAG isomorphism invariant -- that collapses all the equivalent representations into a single canonical form. Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Programming Languages (cs.PL) MSC classes: 68T10 ACM classes: I.2 Cite as: arXiv:2603.21836 [cs.CL]   (or arXiv:2603.21836v1 [cs.CL] for this version)   https://doi.org/10.48550/arXiv.2603.21836 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Ezequiel López-Rubio [view email] [v1] Mon, 23 Mar 2026 11:21:53 UTC (297 KB) Full-text lin...