Boff and Proud
..enoiugh said
11 publicly visible posts • joined 30 May 2013
Reminds me of a model (not AI, pure math) I built many years ago to predict binary choice in pigeons...by random walking parameter values around (what would now be called a sort of evolutionary fit) to find the best fit. N around 100,000. After about a month of trundling around (and a power outage...) it delivered an amazing answer -- on then-common frequentist statistics, the F values were >>1E6, wow, incredibly significant. But then I looked at the proportion of correct predictions -- around 50.2%... My current answer: Don't use frequentist statistics!
My mother recounted a true story (maybe her only true story) of helping out on a relative's farm one Autumn. They got around to making the scrumpy, shovelling up the fallen and rotting apples into some kind of large nonhuman digester. At some point, my mother shouted "Stop! You just shovelled in a dead rat!" Arrgh, the farmer said (this said in vernacular, please), tharrts fine, it gives the scrumpy body...
*I* fondly remember the time when my second slide jammed in the projector, and the idiot running it tried to retrieve it, incidentally dropping all 30 slides on the floor, which he then reinserted in random order and orientation. My talk then became interestingly disjointed. And the time when I was chipping my talk on a soft rock, and the bloody thing shattered into pieces. The more things change the more they stay the same.