The victory of DeepBlue against Garry Kasparov in 1996 left a lasting impression.

The game of Go then became the new battleground: its 19x19 board results in a staggering number of possible combinations, estimated at 10¹⁷⁰!

It took nearly 20 years after DeepBlue’s triumph to witness its successor, AlphaGo, defeat Lee Sedol, the world’s best Go player. This video revisits this incredible achievement and explains the mechanics of the AI through the combination of three methods:

• Monte Carlo Simulation
• Deep Neural Networks (Deep Learning)
• Reinforcement Learning