Automated Symbolic Law Discovery: A Computer Vision Approach

Authors: Hengrui Xing, Ansaf Salleb-Aouissi, Nakul Verma660-668

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We apply our model to a variety of plausible relationships both simulated and from physics and mathematics domains involving different dimensions and constituents. We show that our model is able to identify the underlying operators from data, achieving a high accuracy and AUC (91% and 0.96 on average resp.) for systems with as many as ten independent variables. Our method significantly outperforms the current state of the art in terms of data fitting (R2), discovery rate (recovering the true relationship), and succinctness (output formula complexity).
Researcher Affiliation Academia Hengrui Xing, Ansaf Salleb-Aouissi, Nakul Verma Department of Computer Science, Columbia University, New York, USA h.xing@columbia.edu, ansafsalleb@columbia.edu, verma@cs.columbia.edu
Pseudocode No The paper describes its methodology but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions using the GPLearn software package and provides its URL (1Stephens, T. 2018. GPLearn: Genetic Programming in Python. URL: https://github.com/trevorstephens/gplearn), but it does not state that the authors' own code for De STr OI is publicly available.
Open Datasets No The paper describes generating a 'synthetic dataset' for training and testing, but it does not provide concrete access information (link, DOI, repository, or formal citation) for this dataset or any other publicly available dataset used for training.
Dataset Splits Yes We randomly split the dataset into 80% for training, 10% for early-stopping, and 10% for testing.
Hardware Specification No The paper does not provide specific hardware details (such as GPU or CPU models, processor types, or memory) used for running its experiments.
Software Dependencies No The paper mentions using the 'GPLearn software package' but does not provide a specific version number for it or any other software dependencies like programming languages or libraries.
Experiment Setup Yes We regularize the model with an L2 weight decay and 50% dropouts for the fully connected layers.