Hacker News Reader

Physical/analog computers always suffer from noise limiting their usefulness. So I think it would be natural to apply this to a network architecture that includes noise as an integral Part such as GANs or VAEs.

View on HN

“noise” is integral to all ML systems. You can view this through many lenses, but generalization can be thought of as decoding a noisy signal.

This is true, though what I was getting at was methods that make use of a noise source separate from the input.

Reply on news.ycombinator.com