Generally capable agents emerge from open-ended play

Re:Generally capable agents emerge from open-ended play by Jarwain on Tuesday July 27, 2021 @08:04PM (#27979028)

I've been thinking about an "energy" idea, where actions consume energy but completing goals replenishes it

[ Parent | on HN ]

Re:Generally capable agents emerge from open-ended play by im3w1l on Wednesday July 28, 2021 @04:59AM (#27981715)

I think the issue with

   goals - energy used

is that it requires tuning to make sure that completing goals is better than just idling. A variant that might work better is to optimize "mileage".

    goals / (energy used + epsilon)

Where epsilon has two purposes: it prevents division by zero, and ensures that more goals is (marginally) better than less

[ Parent | on HN ]