Attention for inference compilation

We present a neural network architecture for automatic amortized inference in universal probabilistic programs which improves on the performance of current architectures. Our approach extends inference compilation (IC), a technique which uses deep neural networks to approximate a posterior distribut...

Disgrifiad llawn

Manylion Llyfryddiaeth
Prif Awduron: Harvey, W, Munk, A, Baydin, AG, Bergholm, A, Wood, F
Fformat: Conference item
Iaith:English
Cyhoeddwyd: SciTePress 2022

Eitemau Tebyg