Local Optima in Optimization
The cluster focuses on the common problem in optimization algorithms, particularly gradient descent, where processes get stuck in local maxima or minima instead of reaching the global optimum. Discussions extend to real-world analogies like machine learning, business practices, and human behavior.
Activity Over Time
Top Contributors
Keywords
Sample Comments
Too bad. They're probably chasing local optima that are quite different from the global maximum.
Wouldn't that lead to hitting a local maximum / minimum and getting stuck there?
You're doing it all wrong! You're only optimizing for the local maxima of your optimization's optimization!
Assuming it does not get stuck at some local maximum.
Just because you've never felt the need doesn't mean you aren't stuck in a local minimum.
Not your problem. Just let 'em be stuck in their local optimum (probably better this way) :P
It's very hard to get unstuck from a local maximum
The issue is that you'll hit a local optimum, and all progress will halt.
“Gradient Descent: The Ultimate Optimizer”Did they get stuck in a local optimum?
Isn’t this just saying things seek a local minima (or maxima)?