- September 22, 2020
- Posted by: RSIS Team
- Categories: Engineering, IJRIAS
International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume V, Issue III, January 2020 | ISSN 2454–6186
Identification of the Challenges of Local Minima in Recurrent Neural Networks
Ochonogor Donpaul1, Elebra Charity2 and Friday E.Onuodu3
1Department of Computer Science, School of Postgraduate Studies, Ignatius Ajuru University of Education Rivers State, Nigeria
2&3Department of Computer Science, University of Port Harcourt, Rivers State, Nigeria
Abstract: – Numerous researchers have recently based efforts on the weight of repeated neutral networks through the creation of efficient algorithms, primarily to optimized schemes. For feed forward network, the learning algorithm can become stuck in local minima during gradient descent. This research focuses on recurrent neural networks, local minima in neural networks, optimal learning in the case of feedforward networks, the local minimum is a real question in deeper neural learning and the case of Digital Dividing.
Keywords: Local Minima, Digital Divide, Exploitation, Exploration
