combat1818
New member
- Joined
- Apr 10, 2023
- Messages
- 1
Hi, I've been struggling with this task and would really appreciate some help:
We are given a function f:Rd−>R it is differentiable, L-smooth and non-convex, it has a global minimum x∗We use a version of gradient descent of the form: xt+1=xt−ηt∣∣∇f(xt)∣∣+βt∇f(xt) where ηt,βt>0We want to find ηt,βt which do not depend on the parameter L and guarantee:
T1t=0∑T−1∣∣∇f(xt)∣∣=O~(T2−1)where the O tilde notation ignores logarithmic factors.
We are given a function f:Rd−>R it is differentiable, L-smooth and non-convex, it has a global minimum x∗We use a version of gradient descent of the form: xt+1=xt−ηt∣∣∇f(xt)∣∣+βt∇f(xt) where ηt,βt>0We want to find ηt,βt which do not depend on the parameter L and guarantee:
T1t=0∑T−1∣∣∇f(xt)∣∣=O~(T2−1)where the O tilde notation ignores logarithmic factors.