4 ms·
When I took an ML class we solved that with a temperature parameter, to allow random deviations out of a local minimum. I wonder if there are novel methods now
by Feathercrown 2y ago
When I took an ML class we solved that with a temperature parameter, to allow random deviations out of a local minimum. I wonder if there are novel methods now or if they're just improved versions of temperature?