Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by argonaut on Wednesday March 09, 2016 @04:18AM
(#11251847)
Infinite state/belief/world space. Infinite action space. It's not so much that there aren't rules (there are - physics), it's that the complexity of the full set of rules is exponential or super-exponential.
[ Parent
| on HN ]
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by ogrisel on Wednesday March 09, 2016 @04:45AM
(#11251937)
You can do math with continuous and infinite dimensional spaces.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by argonaut on Wednesday March 09, 2016 @05:19AM
(#11252052)
And? This does not address my argument that the complexity is beyond-combinatorially explosive (infinite spaces). I'm not talking about the space of possible board states. I'm talking about merely the set of all possible actions.
EDIT: clarified my language to address below reply.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by ogrisel on Wednesday March 09, 2016 @03:26PM
(#11255509)
...and it's possible to train learning agents to sense and interact with a world described by high dimensional continuous vector spaces, for instance using conv nets (for sensing audio / video signals) and actor-critic to learn an continuous policy:
http://arxiv.org/abs/1509.02971
The fact that the (reinforcement) learning problem is hard or not is not directly related to whether the observation and action spaces are discrete or continuous.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by ogrisel on Wednesday March 09, 2016 @03:28PM
(#11255514)
see also by the same team:
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by argonaut on Wednesday March 09, 2016 @04:30PM
(#11255872)
There is a near infinite number of such spaces.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by ogrisel on Thursday March 10, 2016 @03:57AM
(#11258035)
I don't understand why would the "number of spaces" matters. What matters is can you design a learning algorithm that performs well in interesting spaces such as:
- discrete spaces such as atari games and go, - continuous spaces such as driving a car, controlling a robot or bid on a ad exchange.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by argonaut on Thursday March 10, 2016 @07:17AM
(#11258721)
A really really large number of distinct decisions that need to be made. A car only needs to control a small set of actions (wheels, engine, a couple others I'm missing). A game player only needs to choose from a small set of actions (e.g. place piece at position X, move left/right/up/down/jump).
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by ogrisel on Thursday March 10, 2016 @08:34AM
(#11259053)
A human brain also has a limited number of muscles to control to interact with the world.
Re:AlphaGo beats the world champion Lee Sedol in first of fi...
by argonaut on Friday March 11, 2016 @07:57AM
(#11266313)
And a much larger set of decisions.