38 thoughts on “AlphaGo & Deep Learning – Computerphile”

  1. It's fascinating that since this was released, human go players have learned so much from alpha go, and vice versa. It's like this interface between human and alpha go is itself an adversarial neural network.

  2. You win a game by taking advantage of the weaknesses of your opponent. Is this real intelligence? I propose it is pseudo-intelligence.

    In game theory and economic theory, a zero-sum game is a mathematical representation of a situation in which each participant's gain or loss of utility is exactly balanced by the losses or gains of the utility of the other participants.
    ‎Zero-sum thinking · ‎Minimax theorem · This is the algorithm Google is running for world control.

    Google world can now win at everything, and you will lose at everything.

    How smart do you think you are enabling this reality?

    If the computer ran a win-win scenario for humanity it would remove this military algorithm by quarantining it.

    I hope someone learns how to think without a military algorithm dominating their logic. This war thinking came from Babylon. Kemet/Egypt ran a win-win scenario in the Ancient times and look what intelligence they had. Cairo is a mathematical engineering wonder yet to be eclipsed. Is this being ignored?

  3. I'm a big Computerphile fan, but this video, unfortunately, had no real information. It also had some aspects that were not true. It kind of reminded me of a USA Today, or Yahoo "News" article. So sadly, this is my first Computerphile complete disappointment.

  4. 2:36 It's funny to say something like "almost infinite" because it's pretty much incoherent, but we know what he means.

  5. no difference in chess. Though go is more complex.
    But alphago also broke chess limits and beat best brute force chess
    You can't check all the consequences of your move in chess

  6. I want to get into Go via using my computer.. anyone have trusted Go game hosts or a direction I should dive into?

  7. On 4:30 they compare to chess implying machine learning is also aplicable to chess. And it is not. Make Deep Learning to try to beat Stockfish and you will see it is imposible without just copying the whole stockfish code. Neural nets are useless on chess. Scientists avoid talk of this implying they have solved also chess with neural nets and this is false. Stockfish, Houdini, Komodo beat any Deep Learning from Google that is not just copying these programmes code and relays only in neural nets. Usually is said how neural net was successful in chess with a program named "Giraffe". Well, that program achieved like 2400 ELO when Stockfish and others are on 3400 ELO. Such Giraffe program would not beat even Grand Masters (ELO 2500) and of course not the top players in 2800 ELO. So if anyone wants to attack these neural net guys just ask them to create a program that beats stockfish and without copying stockfish code of course, just do that teaching of the program playing millions of times against itself and learning and such as they say is successful. Try that in chess and cry viewing that it doesn't work with chess. You hardly can beat a GM and you will never beat Stockfish, Houdini, Komodo with such aproach. Chess continues to be the stone in the foot and the guys know it and run from chess looking for easier pattern games like Go where neural nets work. Go can be a bigger tree but is easier than chess because the patterns locally are very very repetitive, so repetitive that you can just have a book full of them all, you can't do such with chess.

  8. Call me if software will learn to beat every single grand-master in shogi.
    That is, if the government can stil access the satellite network, of course.

  9. even if alphago won the mathes, humans just have to find the weakness in the algorithm, then, computer will always lose, computer doesnt understand concept of left or right, let alone a line. 1st, give the go player a simple calcuator too that just does find some number, but it wont tell the strategy. it is like computer games, no matter how strong your computer made enemies are, they always form a habit. you keep playing against the computer to find the habit, then you say… ah ha, now i know, if i move to the left 2 times and to the right 5 times,, then computer always sends 2 units there x location. then you use that against the computer all the way. computer does a habit, they will never understand the strategy. the flow of computer is a linear .. always a linear, because the current flow is linear.

    let alphago online, until we can find its habit.

  10. actually it was a mix for alpha-go, they want to try starting from random too but alphago had a basis they didn't start from random and went up from there they started from an already somehow set playstyle and went from there.

  11. Should have been more clear on how plotting the line with different values of the line is different from the so called "parameters" in the context of deep learning algorithm . Right ?

  12. No, in chess you can NOT brute force it. At least, not with any computer we have or could conceivably build

  13. it would be really evil to create a video game that learns from the player and becomes unbeatable

  14. Never mind human vs machine Go; I want to see machine Vs machine Go! Like a brainy version of robot wars, with reasearch institutes and companies battling it out for AI glory!

  15. "By the time you're finished, there's no stock market" Hmm, perhaps he used deep learning to predict the fall of capitalism? 😀

  16. AlphaGo started with supervised learning to train the initial network. DeepMind downloaded a database of games performed by human players, and let the network learn from it.
    Afterwards they pitted the network against itself, and letting it improve over its older version.
    And instead of exhaustive search they use probability of which move is most likely to lead to winning.

    The big difference is that AlphaGo uses general AI algorithms, which can be applied to other problems.
    Chess AI like Deep Blue isn't general, it's can only work with Chess.

  17. this is all to do with something we are missing the next other i or square root -1 this is a dimensional shift of biology. Basic maths or physics the reason we dont understand our logic is because intrinsic in our minds is Square root -1 e etc it is the smaller dimesions that exist that allowquantum physics to rule our minds. Personally I think 26 Dimensional time has the amswer and all the constants

  18. Basically that how you learn, the teacher does not give you the answer but tries to help you get to it by yourself keeping you on the right track.

Leave a Reply

Your email address will not be published. Required fields are marked *