The Game of Go has been explored for 2500 years and resulted in a body of theories that Go masters study in Go schools. Alpha Go discovered that human Go understanding was caught in a local optimum, and broke out of it. The game of modern physics started only 130 years ago...
-
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
The game of physics can be stated as POMDP with actions induction (creatively formalizing observations as a model) and deduction (checking the model predictions against reality). The “only” problems: (1) partial observabilty, (2) model evaluation is WAY more costly than Go/Chess.
2 replies 1 retweet 5 likes -
Replying to @mere_mortise @semiDL
The main problems with the current local optima in modeling physics is not lack of data, but finding functions that integrate the data (quantum gravity etc.)
1 reply 0 retweets 2 likes -
This Tweet is unavailable.
-
This Tweet is unavailable.
Ah, now I understand your issue. Approximation with stochastic gradient descent in feedforward networks is popular, but only a special case of machine learning. Any method that can systematically generate and compare computable functions can be used for producing models.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.