High-dimensional Asymptotics of Denoising Autoencoders
–Neural Information Processing Systems
We address the problem of denoising data from a Gaussian mixture using a two-layer non-linear autoencoder with tied weights and a skip connection.
Neural Information Processing Systems
Feb-9-2026, 04:15:14 GMT
- Country:
- Europe
- Germany > Bavaria
- Upper Bavaria > Munich (0.04)
- Switzerland > Vaud
- Lausanne (0.04)
- Germany > Bavaria
- North America > United States (0.14)
- Europe
- Industry:
- Health & Medicine (0.45)
- Technology: