You'll get 50% prediction accuracy for a fair coin no matter what result a model gives. Try testing adversarial games. Mixed strategies make actual sense in them.
replyThe model gives a probability distribution as an answer; you can check to see if two arbitrarily chosen probability distributions are similar or not.
replyTry it yourself to prove you’re useful. Predict my coin flip.
replyI can't, because I don't think you have any coins on you. :)
replyA probabilistic system that cannot reliably reproduce probabilities - somewhat ironic, but absolutely not its purpose.
replyIt’s not really ironic since the goal of the system is to reduce variance. Or I guess ironic in the same way as it is that the Buckingham Palace guards just stand there when they could leave at any time.
replyThat's like not using an LLM for coding assistance because it can't count r's in strawberry.
replydeleted
replyI'm dying to know if you got heads or tails when you flipped the coin.
replydeleted
replyTry asking it to write a program to predict coin flips. That's what it is trained to do.
replyJev literally cannot write programs.
replythis is not what *decision models* do.
reply