Gaurav

In Basu's circus story (PDF), the owner wants to estimate the weight of fifty elephants and suggests weighing Sambo, who looked “average” three years ago, and reporting $50y$. A statistician counters with a contrived sampling plan: choose Sambo with probability $99/100$ and pick any of the other

We often need to adjust a fitted $E[Y|X]$ so that the aggregate $E[Y]$ (or class totals) matches known targets (see here). In binary classification, a single logit shift (or a global temperature) is a monotone transform of the model scores, so if person A had a higher

Textbook bias–variance intuition implies that worst‑case test error should eventually rise as model capacity overtakes the sample size because the variance term in the bound grows while the bias term has already bottomed out (see Vapnik and Chervonenkis, 1971; Bartlett and Mendelson, 2002). Those worst‑case guarantees shrink

Modern applications—from ad platforms calibrating click-through predictions to polling systems incorporating responses to ML algorithms adapting fairness thresholds—share a common challenge: maintaining calibrated weights on live data streams. To address streaming data, we recast raking as a streaming convex optimization problem: minimize the squared error between current weighted