In: Entropy, 2019, vol. 21, no. 8, p. 758
Variational inference with a factorized Gaussian posterior estimate is a widely-used approach for learning parameters and hidden variables. Empirically, a regularizing effect can be observed that is poorly understood. In this work, we show how mean field inference improves generalization by limiting mutual information between learned parameters and the data through noise. We quantify a...
|
In: Mathematische Zeitschrift, 2014, vol. 276, no. 3-4, p. 635-654
|
In: Foundations of Physics, 2009, vol. 39, no. 9, p. 1072-1096
|
In: Rheologica Acta, 2004, vol. 44, no. 2, p. 119-134
|
In: The European Physical Journal B, 2009, vol. 69, no. 4, p. 473-482
|
In: Finance and Stochastics, 2006, vol. 10, no. 4, p. 475-506
|
In: Applied Mathematics & Optimization, 2011, vol. 63, no. 2, p. 217-237
|
In: Journal of Fourier Analysis and Applications, 2007, vol. 13, no. 4, p. 495-510
|
In: Communications in Mathematical Physics, 2013, vol. 320, no. 2, p. 571-601
|
In: Astrophysics and Space Science, 2007, vol. 312, no. 3-4, p. 203-214
|