S
SuperLearn
Courses
AI NotesSoonPlanner
← BackProbability

Lessons

Leftover Must Sum to 1#1 · about 12 minConditioning Zooms the Leftover#2 · about 14 minBayes Reallocates Leftover#3 · about 14 minA Random Variable Names the Leftover#4 · about 12 minExpectation Mixes the Outcomes#5 · about 14 minVariance Is Spread of Leftover#6 · about 14 minDensity Is Leftover per Length#7 · about 11 minJoint Tables Split Leftover Two Ways#8 · about 14 minNamed Families ML Actually Uses#9 · about 13 minIndependence Is a Product#10 · about 12 minMLE Makes the Data Most Probable#11 · about 11 minMAP Spends a Prior on Leftover Belief#12 · about 14 minKL Is Leftover Surprise Between Guesses#13 · about 12 minSoftmax Turns Scores into Leftover#14 · about 11 minAverages Concentrate#15 · about 12 minMarkov Leftover and Nested Expectation#16 · about 12 min
Loading lesson…
Courses›KL Is Leftover Surprise Between Guesses

KL Is Leftover Surprise Between Guesses

DetailedOverviewQuizFlashcard

KL Is Leftover Surprise Between Guesses

Truth 0.8, guess 0.5: entropy 0.5004, cross-entropy 0.6931, KL 0.1927 nats. The reverse is 0.2231. Perfect loss is the entropy, not zero.

This Probability lesson is part of a paid plan. Sign in with a plan that includes this course to read it.