Can You Beat a Computer At Rock-Paper-Scissors?

tekgoblin writes "The New York Times has created a game that uses artificial intelligence to outsmart you. It uses a simple game called Rock-Paper-Scissors which is pretty much known by everyone on the planet by now. The computer tries to mimic human reasoning by building on simple rules and statistical averages. So based on the rules of the game and your previous moves, the computer tries to make predictions on your next move. The game has 2 modes, the first being Novice, where the computer learns the game from scratch, and Veteran, where the computer has experience of over 200,000 rounds of previous experience."

Re:This game is random , you can't outsmart someon

[Tukz]

Your choices aren't truly random though.
It's been argued many times, that people make choices in patterns.

Re:Best strategy

[N1AK]

You may have won in a specific case, that doesn't mean it is a winning strategy. Mathematically, a strategy based on pure randomness can't be more or less likely to win on average. Why? Because you can effectively ignore what the non-random player selects. There is a 1/3 chance you will randomly pick the same, 1/3 that you will randomly pick the one that beats them and 1/3 that you will pick the one that loses.

Your theory about patterns is wrong. Even if they are incorrectly detecting a pattern it doesn't change the odds of your random choice winning/losing/drawing.

A good strategy would mix random choices with selectively picked moves. Effectively you would need to double-guess what the computer system thinks your pattern is. Very good systems would then track if they are being tracked etc. Two 'perfect' systems would trade increasingly rare attempts to score, as they realise that the best reliable result they can hope for is a draw. This is because any winnning strategy must be based on predicting your opponents choices, the more you act upon your predictions the more a good opponent can learn about how your algorithm works and how to defeat it.