Differences

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

--- nlp:hmm [2021/01/16 02:40] – [Recent Advances] jmflanig
+++ nlp:hmm [2023/06/15 07:36] (current) – external edit 127.0.0.1
@@ Line 1: / Line 1: @@
 ====== Hidden Markov Models ======
+===== Basics =====
+  * HMMs vs CRFs: See [[https://cs.stanford.edu/~pliang/papers/asymptotics-icml2008.pdf|Liang & Jordan 2008 - An Asymptotic Analysis of Generative, Discriminative, and
+Pseudolikelihood Estimators]] especially table 2, and summary [[http://www.cs.cmu.edu/~mgormley/courses/10418/slides/lecture10-crf.pdf|slide 44 here]]
+===== Applications in MT =====
+  * [[https://www.aclweb.org/anthology/C96-2141.pdf|Vogel et al 1996 - HMM-Based Word Alignment in Statistical Translation]]
+  * [[https://www.aclweb.org/anthology/P18-2060.pdf|Wang et al 2018 - Neural Hidden Markov Model for Machine Translation]]
 ===== Recent Advances =====
   * [[https://www.aclweb.org/anthology/2020.emnlp-main.103.pdf|Chiu & Rush 2020 - Scaling Hidden Markov Language Models]] Scaling the number of hidden states to 2^15 states, with exact inference.
   * This paper shows that HMMs can be viewed as a special case of RNNs: [[https://openreview.net/pdf?id=rJxEso0osm|Buys et al 2018 - Bridging HMMs and RNNs through Architectural Transformations]]
+===== Tutorials and Introductions =====
+  * [[http://www8.cs.umu.se/kurser/5DV122/HT13/material/Noah_Smith_HMMs.pdf|Smith 2004 - Hidden Markov Models:
+All the Glorious Gory Details]]
+===== Related Pages =====
+  * [[ml:Conditional Random Field]]
+  * [[ml:Graphical Models]]