Re:Beyond A*: Better Planning with Transformers
by SnowflakeOnIce on Friday February 23, 2024 @02:57PM
(#39485351)
How would one go about proving that a learned heuristic (something from an AI model) is in fact admissible?
[ Parent
| on HN ]
Re:Beyond A*: Better Planning with Transformers
by jvanderbot on Friday February 23, 2024 @04:29PM
(#39486349)
For something like focal search, it doesn't even need admissibility, you just apply it as a second selection heuristic among the choices your admissable heuristic returns as 'top k' choices.
Re:Beyond A*: Better Planning with Transformers
by SnowflakeOnIce on Friday February 23, 2024 @05:01PM
(#39486672)
So a tiebreaker then?