Skip to content
Kudos AI

Optimization for Learning

Every model on this site is fitted by the same loop: look at the slope, take a step. What decides whether that loop converges in forty steps or diverges in three is not the model - it is curvature, noise, and the size of the step. All three are measurable before the first epoch runs.

Intermediate300 XP~1 h90% to advance

Sign in to take quizzes, earn XP, and unlock stages as you reach 90% mastery.

  1. The Step, and the Edge of Stability

    25 min · 100 XP

    What a gradient step is minimising, why a minibatch gradient is the full gradient plus noise rather than a different gradient, and the exact step size above which descent stops descending.

    Open lesson →

    Sign in to take the 3-question quiz.

  2. Curvature, and What Momentum Buys

    25 min · 100 XP

    The condition number is the single number that predicts how slow plain descent will be, and momentum changes that prediction from κ to √κ - a claim precise enough to check to six decimal places.

    Open lesson →

    Sign in to take the 3-question quiz.

  3. Noise, Schedules, and Adaptive Methods

    25 min · 100 XP

    Stochastic gradient descent with a fixed step does not converge - it settles into a ball whose radius grows as √η. Decaying the step removes the floor; Adam rescales each coordinate; and on a long enough run the fastest method is the one that finishes last.

    Open lesson →

    Sign in to take the 3-question quiz.

Complete all modules → earn the Optimization for Learning badge