SORT 2025

Introduction

Introduction

Univar: + Solomonoff’s theory of inductive inference. + How to form prior for bayes, computation complexity?

stochastic gradient descent somewhere + algorithms which use stats generally?

bounded-error probabilistic polynomial (BPP) + P is special case of BPP. don’t use randomness + have access to random tape + required to be correct 2/3 of time

Zero-error probabilistic polynomial (ZPP)

Randomised polynomial (RP) time + if returns yes then definitiely yes, not case for no + Co-RP exists too

PP probabilistic polynomial + like BPP but error < 1/2 rather than < 1/3

post selection and BPP_path

Jeffreys divergence (J-divergence)

The Kolmogorov–Smirnov test (K-S test)

Conjugate priors: If the prior \(P(\theta)\) and the posterior \(P(\theta | X)\) are in the same family of distributions (eg both Gaussian), then the prior and posterior are conjugate distributions

entropy: use \(E_i\) \(E_j\) language

in events, stuff on marginaliation, discrete continuous to variables h3

simple continous distribution split out: what is motivation, how do they arrive? this should indicate where is pdf they belong + for ones not taken out, move to h3 called probability to sort, page for each one

exploratory data analysis

in page on "testing population means", page on type 1, type 2 errors somewhere.null hypothesis, power of a test