Profile cover photo
Profile photo
Mustafa Muhammad
20 followers
20 followers
About
Mustafa's interests
View all
Mustafa's posts

Post has attachment

Post has attachment

Post has attachment

Post has attachment

Post has attachment
"The approach mimics how young children learn between the ages of one and four years old, and is essentially helping the robots to develop reliable hand-eye coordination."

Post has attachment

Post has attachment
"Go includes about 1,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000, 000,000,000,000,000 possible positions to choose from, and the game can't be won with "simple" brute-force search techniques (with current amounts of processing power, anyway).
Instead, AlphaGo uses a novel mix of deep neural network machine learning and tree search techniques. Without getting into the complexities of it, DeepMind taught AlphaGo how to play the game by watching human Go masters play. The system then honed its strategy by pitching instances of AlphaGo against each other in a process known as reinforcement learning. Thousands (or millions?) of iterations later (with a bit of AI secret sauce), and AlphaGo was ready to, er, go."

Post has attachment

Post has attachment

Post has attachment
Wait while more posts are being loaded