What if we get local maxima when we choose w, b in gradient descent algorithm

[Jashwanth_Reddy]

I think the algorithm fails if we guess w , b at local maxima

[tharunnayak14]

Gradient Descent starting at the "top of the hill"

Always exciting to revisit these fundamental concepts as we look at things differently. I was curious about the starting point for Gradient Descent when you are on top of the hill as explained in the video. How do you begin taking that “baby step” with the fastest drop when you are starting on top of the hill and all immediate directions are equal?

refer this link

[TMosh]

For most simple regression systems, the cost function is known to be convex. So a local maximum is not possible.

[Yahyaa1]

May I encounter cost function in other models that may cause the gradient descent algo to converge to a local maxima?
How to prevent it in that case?
Thx.

[TMosh]

If any of the initial weight values cause the gradients to be non-zero, you won’t start at a maximum, and the gradients will lead away from the maximum on the first iteration.

You would have to be extremely unlucky to choose initial weight values that had all of the gradients at exactly zero.

In practice, you can safely ignore this as a potential issue.