nilinswap / neuro-evolution

A project on improving Neural Networks performance by using Genetic Algorithms.

Multi-Objective Neuro Evolution

Abstract

In this project we implement a memetic algorithm, in that, we are training neural networks using evolutionary algorithms and occasionally apply back-propagation on cluster-heads we recieved after applying clustering. We used NEAT techniques to realise Neural Network and combined it with NSGA-II to implement Genetic algorithm. We have also done experiments on Transfer learning using this model. See branch 'dataset3_tl' for that.

Contributors

1. Swapnil Sharma
2. Aditya Pimparkar
3. Atul Kumar Sinha
4. Kunal Gupta
5. Raghav Khandelwal