domokato - 22 February 2012 12:07 PM

VYAZMA - 22 February 2012 11:25 AM

I played 100 rounds. Veteran.  I could easily begin to see the pattern that the computer was using to beat me.  I threw 10 consecutive rocks.  By the 3rd time the computer faithfully threw out paper every time.  That’s not random. It’s complicated math that most humans are incapable of doing.
26 wins 22 ties 52 losses.

That’s interesting. It doesn’t say anything about considering the current opponent’s play style when it’s thinking, although it totally should.

In fact…since it doesn’t say anything about learning from the current player, you should be able to play certain patterns of 5 throws or so over and over again and beat it pretty regularly. Until your games become a statistically significant portion of its considerations lol

FreeInKy - 22 February 2012 09:22 AM

I refuse to participate in this improper game that does not include either lizard or Spock.

FreeInky nailed it.  Besides, if the computer were at all intelligent it would always choose Spock. 

Occam

From the web page:

A truly random game of rock-paper-scissors would result in a statistical tie with each player winning, tying and losing one-third of the time. However, people are not truly random and thus can be studied and analyzed. While this computer won’t win all rounds, over time it can exploit a person’s tendencies and patterns to gain an advantage over its opponent.

As Mingy Jongo said, it is really nearly impossible for humans to produce random series. So George’s strategy is not bad. But without analyzing what the other is doing, or not being able to do it because the other plays really random, one cannot hope effectively to beat the other in longer games of many rounds. So trying to find a pattern in the opponents moves is the only way to do better, even if it fails against a random player. See the contest, I linked before. I expect you to see that the best program at the moment will beat you when you do not use an external randomiser (like George did). The best program (bayes14) has already proven to do better against the other programs.

As additional funny story: a colleague of mine told me in his school days he faked a lot of absences in his ‘absence book’. The math teacher did a statistical analysis on it, and could show that the absences were faked.

In the program in the OP, the ‘veteran’ expects you to play as an average player. So your best chance is being not average. In novice mode the program analyses you. The better you can simulate a random series, the better you can do, but really beating the program, way from the 50/50 is then very difficult.

GdB - 23 February 2012 12:10 AM

From the web page:

A truly random game of rock-paper-scissors would result in a statistical tie with each player winning, tying and losing one-third of the time. However, people are not truly random and thus can be studied and analyzed. While this computer won’t win all rounds, over time it can exploit a person’s tendencies and patterns to gain an advantage over its opponent.

As Mingy Jongo said, it is really nearly impossible for humans to produce random series. So George’s strategy is not bad. But without analyzing what the other is doing, or not being able to do it because the other plays really random, one cannot hope effectively to beat the other in longer games of many rounds. So trying to find a pattern in the opponents moves is the only way to do better, even if it fails against a random player. See the contest, I linked before. I expect you to see that the best program at the moment will beat you when you do not use an external randomiser (like George did). The best program (bayes14) has already proven to do better against the other programs.

As additional funny story: a colleague of mine told me in his school days he faked a lot of absences in his ‘absence book’. The math teacher did a statistical analysis on it, and could show that the absences were faked.

In the program in the OP, the ‘veteran’ expects you to play as an average player. So your best chance is being not average. In novice mode the program analyses you. The better you can simulate a random series, the better you can do, but really beating the program, way from the 50/50 is then very difficult.

It’s very interesting isn’t it.

I had a go last night on novice. Not got my score but was well behind and wasn’t aware of playing to any pattern.

Stephen

wins   draws   loses

19     16     15

Against veteran.

I was trying to select what I was least comfortable with, thinking that would cause me to play against what my brain was working out as the best thing to do.

I only didn’t carry on doing this when I’d had a couple of losses with the same thing, as the computer does seem to be programmed to keep winning if you keep going, as others have noticed.

Stephen

Against novice: 13 - 12 - 5 with this random generator.

Second novice round: 10 - 10 - 10.

Against veteran: 8 - 14 - 8.

Made new random series everytime.

I found a very interesting website detailing an even simpler game than RPS called “matching pennies”.  It includes videos about the game and AI sequence prediction in general:
http://matchingpennies.com

Mingy Jongo - 23 February 2012 10:01 AM

I found a very interesting website detailing an even simpler game than RPS called “matching pennies”.  It includes videos about the game and AI sequence prediction in general:
http://matchingpennies.com

I’ve won…

I’ll make a dump of my brain, and send it in. Would my brain fit in 20MB?

FreeInKy - 22 February 2012 09:22 AM

I refuse to participate in this improper game that does not include either lizard or Spock. Fail.

Facebook use has compelled me to reach for the “like” button under this comment.