Who is Afraid of Big Bad Minima? Analysis of gradient-flow in spiked matrix-tensor models

Stefano Sarao Mannelli, Giulio Biroli, Chiara Cammarota, Florent Krzakala, Lenka Zdeborová

Neural Information Processing Systems 

When the signal-to-noise ratio is large enough, the success of gradient descent can thus be understood by a trivialisation transition in the loss landscape: either there is only a single minima, or all minima become "good", and no

Similar Docs  Excel Report  more

TitleSimilaritySource
None found