0.25
0.5
0.75
1.25
1.5
1.75
2
Infer.NET - Practical Implementation Issues and a Comparison of Approximation Techniques
Published on Dec 31, 200710029 Views
Infer.NET is an efficient, general-purpose inference engine developed at Microsoft Cambridge by Tom Minka, John Winn and others. It aims to be highly efficient, general purpose and extensible --- thre
Related categories
Chapter list
Expectation Propagation & Variational Message Passing00:00
Overview00:55
Divergence minimisation - 101:48
Divergence minimisation - 204:03
Message passing07:10
Overview10:07
Beta‐Bernoulli10:22
Gaussian with fixed variance13:25
Surprising conclusion14:51
Demo #115:33
Learning mean and precision - 121:29
Learning mean and precision - 222:07
Learning mean and precision - 322:35
Learning mean and precision - 423:10
So...24:30
Demo #224:56
Improper messages are necessary27:30
Improper message ‘fixes’28:29
Product of Gaussians - 137:27
Product of Gaussians - 238:22
Product of Gaussians - 339:47
Exact posterior for A40:46
Posterior for A vs. variance of B41:08
Posterior for positive A vs. variance of B43:42
Demo #345:14
Summary45:25
Thanks!45:34