Extreme learning machine with sparse connections
Zuo Bai · 2015
with state-of-the-art result on ETH-80.In summary, this thesis studies ELM with sparse connections.We first present the sparse ELM as an alternative solution for the unified ELM, which significantly reduces the storage space and testing time.Additionally, the sparse ELM has better scalability, making it preferable to handle large-scale applications.Later, we conduct some research on the ELM-LRF and shows that it is especially suitable and efficient for highly correlated applications, such as image processing, speech recognition, natural language processing, etc.