Tag Archives: math

On forecasting

I’m really enjoying Nate Silver’s The Signal and the Noise, particularly as someone who has come to appreciate statistics and machine learning only grudgingly and late in life. This book would have been a much gentler introduction to the concepts of statistical prediction than my own: the dense, equation-filled Pattern Recognition and Machine Learning by Bishop. However, my edition of Signal has a few errors that had me scratching my head, such as a chessboard showing an impossible opening position. And there was this gem:

“At NASA, I finally realized that the definition of rocket science is using relatively simple psychics to solve complex problems,” Rood told me.

I guess ML would be easier if we had more of these simple psychics to go around.