Differences

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

--- ml:em_algorithm [2021/03/30 10:28] – created jmflanig
+++ ml:em_algorithm [2023/06/15 07:36] (current) – external edit 127.0.0.1
@@ Line 1: / Line 1: @@
 ====== EM Algorithm ======
+===== Older Papers =====
+  * Hard EM (also called "Viterbi training" or "sparse EM")
+    * [[https://www.cs.toronto.edu/~hinton/absps/emk.pdf|Neal & Hinton 1998 - A view of the EM algorithm that justifies incremental, sparse, and other variants]] Calls it "sparse EM"
+    * [[https://aclanthology.org/W10-2902.pdf|Spitkovsky et al 2010 - Viterbi Training Improves Unsupervised Dependency Parsing]]
+    * [[https://aclanthology.org/P10-1152.pdf|Cohen & Smith 2010 - Viterbi Training for PCFGs: Hardness Results and Competitiveness of Uniform Initialization]]
 ===== Recent Papers =====
-  * [[https://www.aclweb.org/anthology/2020.tacl-1.15.pdf|Nishida & Nakayama - Unsupervised discourse constituency parsing using Viterbi EM]]
+  * [[https://www.aclweb.org/anthology/2020.tacl-1.15.pdf|Nishida & Nakayama 2020 - Unsupervised discourse constituency parsing using Viterbi EM]]
+  * [[https://arxiv.org/pdf/2209.01232.pdf|Wang et al 2022 - Elaboration-Generating Commonsense Question Answering at Scale]] Uses an EM-style algorithm to train the background knowledge generator
+===== Related Pages =====
+  * [[nlp:Unsupervised Methods]]