What is different about AlphaGo versus Deep Blue? AlphaGo is a learning system and a hybrid system

AI expert Ben Goertzel explains AlphaGo.

We are using learning now and not just simplistic search

Alphago is not just one system

  1. Deep Learning neural networks
  2. Monte Carlo Sampling
  3. Game tree search

Goertzel at Hanson Robotics is using hybrid architecture as well
Deep Learning
Symbolic Reasoning
Computational Linguistics

Machine Translation- Google is using pure statistical methods but in the literature there is hybridization

AlphaGo is a narrow AI achievement. But it is a manifestation of broader trends which will bring us Artificial General Intelligence.

You can simulate millions of games a day at high speed but we cannot simulate the world at high speed.
Monte Carlo sampling only works with things that are game like and more contained.
However, you could simulate imprecisely or approximately.
With more sensing (widespread imaging) and trillions of sensors that could help get us closer to having information for simulations but that would be far beyond the 19 X 19 Go board.

Differences over old neural networks.
Layers of specialization. simple to complex.

About The Author

Add comment

E-mail is already registered on the site. Please use the Login form or enter another.

You entered an incorrect username or password

Sorry, you must be logged in to post a comment.