Named after Johannes Kepler, the tool ingests CSV or TSV files from standard input and identifies compact algebraic relationships linking features to a target measurement. It utilizes best-first recursive reduction with invertible transformations like logarithms and arcsine functions to isolate underlying patterns. Candidate laws are output as JSON alongside r-squared values, with fitted constants automatically rounding to rational numbers or known mathematical constants. Evaluations in the standard SRBench suite demonstrate strong ground-truth recovery speeds across both the Feynman and Strogatz benchmark suites.
Key points
The system delivers exact algebraic relationships in seconds or milliseconds directly fro…: The system delivers exact algebraic relationships in seconds or milliseconds directly from standard input tabular data.
Robust performance is maintained under low-noise settings, recovering 56% of Feynman equa…: Robust performance is maintained under low-noise settings, recovering 56% of Feynman equations and 42.9% of Strogatz problems at 0.001 noise.
Sources checked by our analysis pipeline: 1
Symbolic regression in seconds.
Keplearn is named after Johannes Kepler, who deduced his laws of planetary motion by studying tables of astronomical observations taken by Tycho Brahe. That is what this program attempts to do: given columns of measurements, find the compact law that relates them.
More formally, it's a deterministic symbolic regression engine (written in Rust) that recovers closed-form equations from tabular data by best-first recursive reduction.
- Reads a TSV or CSV from stdin, prints candidate equations as JSON
- Single binary, zero dependencies (fully static with the musl target)
- Designed to answer in seconds: exact laws usually resolve in
milliseconds, and
--timeout
caps the hard cases - Deterministic; the same input produces the same output, byte for byte
Computational researchers have long relied on standard FP32 precision to prevent trajectory instability and maintain physical accuracy in molecular simulations, leaving specialized GPU hardware underutilized. By splitting full-precision operands into lower-precision components, new techniques leverage Tensor Core matrix-multiplication throughput without degrading physical fidelity. This approach delivers substantial inference acceleration while maintaining strict quantum-mechanical conservation laws without requir…
Meta has integrated an automated testing architecture into its live development environment to catch regressions across its massive codebase before changes are merged. The system uses large language models to construct executable, failing tests that reveal unexpected behavioral shifts introduced by newly submitted code updates. By combining algorithmic filtering with focused engineer oversight, the initiative aims to safeguard massive systems without overwhelming developers with unnecessary alerts.
The release of the 120-billion-parameter Nemotron 3.5 Super VL brings advanced multimodal intelligence and long-context processing directly to enterprise cloud environments. By open-sourcing the model weights, datasets, and post-training recipes, developers can tailor the system to internal workflows while hosting it securely on their own clusters. This setup enables high-throughput agentic tasks such as cybersecurity analysis and automated coding with reduced operational overhead.