DEVDAN: Deep Evolving Denoising Autoencoder

The Denoising Autoencoder (DAE) enhances the flexibility of the data stream method in exploiting unlabeled samples. Nonetheless, the feasibility of DAE for data stream analytic deserves an in-depth study because it characterizes a fixed network capacity that cannot adapt to rapidly changing environments... (read more)

Results in Papers With Code
(↓ scroll down to see all results)