luca
@fahrenheit
ML researcher, interested in math and probabilistic machine reasoning.
luca
The Fisher information matrix is simply the Hessian of KL divergence at the point where two distributions are equal.
Metacademy - Differential geometry for machine learning
moments are important theoretically because they provide an alternative way to fully and uniquely specify a probability distribution, a fact that is intuitive if you understand how moment’s quantify a distribution’s location, spread, and shape
Understanding Moments
This is so good.
Differential geometry is all about constructing things which are independent of the representation
