Feynman Symbolic Regression Dataset for Scientific Discovery

Designed for evaluating Symbolic Regression methods in the context of Scientific Discovery. It is derived from the Feynman Symbolic Regression Database, with formulas and variables reviewed to define realistic sampling ranges. The dataset is categorized as the 'Medium' set within the SRSD-Feynman collection.

Use Cases

• Benchmark Symbolic Regression algorithms on formulas from the Feynman Symbolic Regression Database.
• Evaluate the potential of SR methods to rediscover physical laws using the defined variable sampling ranges.
• Analyze the performance of SR for Scientific Discovery (SRSD) on the 'Medium' difficulty subset of formulas.

Strengths

• Derived from the established Feynman Symbolic Regression Database.
• Formulas and variables were reviewed to define realistic sampling ranges for evaluation.
• Part of a collection specifically designed for Symbolic Regression for Scientific Discovery (SRSD).

Limitations

• The specific number of rows, columns, and file size are unknown.
• The exact formulas, variables, and data format are not detailed in the provided input.
• The 'Medium' set's difficulty and scope relative to other sets is not quantified.

Provenance

Source

Feynman Symbolic Regression Database, via Hugging Face.

Collection Method

Formulas and variables were reviewed to design realistic sampling ranges for evaluation.

Freshness

Last updated on March 5, 2024.