This repository contains code related to the research paper entitled "The Limitation of neural nets for Approximation and optimization"
In this paper, we explore the effectiveness of neural networks as surrogate models to approximate and minimize objective functions in optimization problems. The study involves the comparison of various activation functions, evaluation of surrogate models on different test problem sets, and an analysis of their impact on optimization algorithms.
Clone the repository and run the specific Python file to reproduce the experiments and results presented in the paper.
git clone <repository_url>
cd <repository_name>
# Run experiments (i.e. for section 2)
python section2_exp.py