CLT rates for the Fisher information matrix of Gaussian scale mixtures

Author

Lampros Gavalakis

Link:  CLT rates for the Fisher information matrix of Gaussian scale mixtures

This blog explores convexity properties of information functionals for mixtures of centered Gaussian random variables (with respect to the variance)

Open questions:

• The CLT rates above are for i.i.d. Gaussian mixtures. Can we extend it to weakly
  dependent mixtures like the neural network output in Figure 1?
• Are the rates sharp?

For more information of Lampros’s work read the bio or get in touch