This project focuses on the development and optimization of Extreme Learning Machine (ELM) neural networks. The aim is to enhance the model's fitness and precision by implementing different bio-inspired algorithms between the input and hidden layer. These algorithms are used to optimize the weights and biases, as well as perform feature selection.
The project implements three different bio-inspired algorithms:
- Genetic Algorithm (GA) 🧬
- Particle Swarm Optimization (PSO) 🐦
- Coral Reef Optimization (CRO) 🐠
The CRO-ELM is a novel approach proposed in this project, aiming to improve the current state-of-the-art.
Per the author and tutor's instructions, the implementation of the neural networks and the formulas must be done from scratch. The use of any external libraries like Tensorflow or Keras for these implementations is strictly prohibited.
You can access the final project report here.
For more details about the project, please refer to the Final Presentation.
This project serves as a stepping stone towards more advanced and efficient bio-inspired algorithms for neural network training. Future work will focus on refining these algorithms and exploring their applications in various domains.