Differences

This shows you the differences between two versions of the page.

--- data_mining:neural_network:model_combination [2017/04/01 13:20] – [Approximating full Bayesian learning in a NN] phreazer
+++ data_mining:neural_network:model_combination [2017/08/19 20:12] (current) – [Approximating full Bayesian learning in a NN] phreazer
@@ Line 74: / Line 74: @@
 Sample weight vectors $p(W_i|D)$.
+In Backpropagation, we keep moving weights in the direction that decreases the costs.
+With sampling: Add some gaussion noise to weight vector, after each update.
+Markov Chain Monte Carlo:
+If we use just the right amount of noise, and if we let thei weight vector wander around for long enough before we take a sample, we will get an ubiased sample form the true posterior over weight vectors.
+More complicated and effective methods than MCMC method: Don't need to wander the space long.
+If we compute gradient of cost function on a **random mini-batch**, we will get an unbiased estimate with sampling noise.
+====== Dropout ======
+See [[data_mining:neural_network:regularization|Regularization]]