http://www.scmp.com/tech/apps-gaming/article/2059568/winning-online-go-master-revealed-be-ai-programme-deepmind
This is a significant step forward for AI. Weiqi is far more complex than international chess and the heuristic process involved in playing weiqi is far more advanced than international chess. There is no way for computer to analyse all the permutations because it exceeds the computing capacity. Instead, this advance Alphago is interpreting weiqi based on logic which is simply amazing.
In our previous posting about the new social structure, it looks more probable that the future is coming earlier than thought with the advancement of AI.
Subscribe to:
Post Comments (Atom)
2 comments:
This new heuristic process is revolutionary as computer in the past was programmed to analyze all the permutations before solving any problem and if the permutations exceeded the computer capacity, there was no way to solve the problem.
There are many problems which exceed the computer capacity and this new heuristic process can be applied to solve those problems now.
The first Alphago was uploaded with approximately 1m weiqi game moves for it to analyse the game which would explain the less spectacular winning because weiqi had more than 1m game moves. The upgraded and advanced Alphago is not loaded with any game moves but only with game rules and game logic. Therefore, the new Alphago heuristic process is really remarkable.
Post a Comment