Can You Beat the Rock-Paper-Scissors Supercomputer?
- Thread starter Andy Chalk
- Start date
Recommended Videos
i won 11 tied 10 and lost 9.
so i beat the supercomputer, my computer then promptly froze and i had to restart it.
i may have angered our future robotic overlords...
so i beat the supercomputer, my computer then promptly froze and i had to restart it.
i may have angered our future robotic overlords...
It works by predicting what an average human would do. So all you have to do to win is think like a retard.
So whenever you think "I picked scissors the last three turns, maybe I should do rock now" you should do scissors again, instead.
I got 18-13 on veteran.
So whenever you think "I picked scissors the last three turns, maybe I should do rock now" you should do scissors again, instead.
I got 18-13 on veteran.
Long term it would be 1/3 across the board. If you want to get fancy, it wouldn't be exactly 1/3 but in a small range that mathematicians, economists, and psychologists would accept as 1/3. Just like if you flip a coin 10 million times, you will get close to 5 million heads. The only way the "truly random" would come into play was if you were allowed to make ANY symbol with no rules on that. Statistics are your friend.Ben Simon said:
After 45 rounds on veteran, I somehow got 21 wins, 15 draws and 9 losses. 
how does that work out? (Because I'ma American >:O jk)
how does that work out? (Because I'ma American >:O jk)
10-5-7. It was definately learning my patterns though, I had to change my style at least five times in that short period to keep on top of it. Probably one of the most difficult opponents I've ever faced.
Interesting post.
Edit: I tried to face it like I would a human, over the long term it will certainly beat any opponent, because it has date, it's like playing blackjack. a sound strategy will cut your losses but the house will win long term. The only way to keep on top of the machine is to keep changing your stragegy, and remember that it is going off your own thought processes, so to beat it, beat yourself.
Interesting post.
Edit: I tried to face it like I would a human, over the long term it will certainly beat any opponent, because it has date, it's like playing blackjack. a sound strategy will cut your losses but the house will win long term. The only way to keep on top of the machine is to keep changing your stragegy, and remember that it is going off your own thought processes, so to beat it, beat yourself.
I did a lot better against veteran than I did on novice. Apparently there's some method to my madness that is Schordinger Fu that he can trace.
I got 9-13-20 when I played against it.
On Novice.
My usual strategy of, "play rock, then scissors, then paper," didn't work.
On Novice.
My usual strategy of, "play rock, then scissors, then paper," didn't work.
Interesting. Because of the nature of statistics, the more games played, the more likely it is to reach that perfect percentage. But, it is just statistics, so you could play a thousand games of it and win everyone. In veteran, what is the use of pulling data from thousands of games played? No matter what you input, there is only three possible choices. I saw a comment earlier that says the easy way to beat the computer is to pick all one. That is true and fairly obvious. The more you pick it, statistically speaking, the more likely you are not to pick it the next round. Only, if you are deciding to pick the same thing every time, the computer is believing in the random statistical chance, and you are picking a single choice 100% of the time.
Also, the concept of random is fairly interesting when it comes to computers. When I was in the 8th grade, I took an advanced class that introduced us to fairly complex computer concepts. In the class, we built computers, learned basic, created a basic program that let us control a robot though it, stuff like that. I know basic isn't complicated, but this was also before a lot of modern languages which are even easier in many instances. I am 30 now, so you do the math as to when that was. I don't know if things have changed, but here is my story. With my teacher, I synchronized all the computers in the room to the exact same time. Then we used a simple randomize timer statement with a few other simple commands, and we made a tiny tiny little program. Basically, we hit space bar and it would spit out a random number. Here is the kicker, if we synchronized the button press, it would give us the exact same digit because as it turns out, the so called random number was determined in part by the time and when in that time the button was pressed. I was just wondering if anything has changed?
Also, the concept of random is fairly interesting when it comes to computers. When I was in the 8th grade, I took an advanced class that introduced us to fairly complex computer concepts. In the class, we built computers, learned basic, created a basic program that let us control a robot though it, stuff like that. I know basic isn't complicated, but this was also before a lot of modern languages which are even easier in many instances. I am 30 now, so you do the math as to when that was. I don't know if things have changed, but here is my story. With my teacher, I synchronized all the computers in the room to the exact same time. Then we used a simple randomize timer statement with a few other simple commands, and we made a tiny tiny little program. Basically, we hit space bar and it would spit out a random number. Here is the kicker, if we synchronized the button press, it would give us the exact same digit because as it turns out, the so called random number was determined in part by the time and when in that time the button was pressed. I was just wondering if anything has changed?
Heh. Quite cute. After awhile I picked up it's veteran strategy and was able to tie consistantly to test my theory out.
Well, I either tied(more likely) or won.
Well, I either tied(more likely) or won.