2 ms·
There was a Chinese Legend that the best person in Go is not one who beat every single opponents, but in absolute control, which means he could play the game an
by ksec 7y ago
There was a Chinese Legend that the best person in Go is not one who beat every single opponents, but in absolute control, which means he could play the game and always end with a draw.
I certainly think it was more of a myth, as it was used in context to play Go with the Emperor, and always play to a draw or lose by a stone. ( You cannot or not allowed to win against the Emperor )
I wonder if this is now possible to achieve with AI.
And why do they not move on to AlphaZero, which was better than AlphaGo Lee played against, and when AlphaGo had two stone advantage, Surely they can now test with level of playing field? And next step to only allowed AI use a specific amount of energy, or the energy an average human brain consume?
I mean right now my guess is that DeepMind is likely using thousands if not millions times more energy to make those moves, that is like an Army of million against one General.
And what real world productivity can we adopt and use from DeepMind?