Visual Learning and Recognition: Difference between revisions

Visual Learning and Recognition (view source)

364 bytes added , 19 November 2020

5,337

edits

@@ Line 1,056: / Line 1,056: @@
 ==GANs and VAEs==
+===Pixel-RNN/CNN===
+* Fully-visible belief network
+* Each pixel depends on it's adjacent pixels
+* Training:
+** Decompose likelihood
+** <math>P_{\theta}(x) = \prod_{i=1}^{n} P_{\theta}(x_i | x_1, ..., x_{i-1})</math>
+;Pros:
+* Can explicitly compute P(x)
+* Explicit P(x) gives good evaluation metric
+;Cons:
+* Sequence generation is slow
+* Optimizing P(x) is hard.
 ==Will be on the exam==