The sequence memoizer

The sequence memoizer The Sequence Memoizer By Frank Wood, Jan Gasthaus, Cédric Archambeau, Lancelot James, and Yee Whye Teh Abstract Probabilistic models of sequences play a central role in most machine translation, automated speech recognition, lossless compression, spell-checking, and gene identification applications to name but a few. Unfortunately, realworld sequence data often exhibit long range dependencies which can only be captured by computationally challenging, complex models. Sequence data arising from natural processes also often exhibits power-law properties, yet common sequence models do not capture such properties. The sequence memoizer is a new hierarchical Bayesian model for discrete sequence data that captures long range dependencies and power-law characteristics, while remaining computationally attractive. Its utility as a language model and general purpose lossless compressor is demonstrated. 1. intRoDuction It is an age-old quest to predict what comes next in sequences. Fortunes have been made and lost on the success and failure of such predictions. Heads or tails? Will the stock market go up by 5% tomorrow? Is the next card drawn from the deck going to be an ace? Does a particular sequence of nucleotides appear more often then usual in a DNA sequence? In a sentence, is the word that follows the http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Communications of the ACM Association for Computing Machinery

Loading next page...
 
/lp/association-for-computing-machinery/the-sequence-memoizer-O0uprcmqxa
Publisher
Association for Computing Machinery
Copyright
Copyright © 2011 by ACM Inc.
ISSN
0001-0782
D.O.I.
10.1145/1897816.1897842
Publisher site
See Article on Publisher Site

Abstract

The Sequence Memoizer By Frank Wood, Jan Gasthaus, Cédric Archambeau, Lancelot James, and Yee Whye Teh Abstract Probabilistic models of sequences play a central role in most machine translation, automated speech recognition, lossless compression, spell-checking, and gene identification applications to name but a few. Unfortunately, realworld sequence data often exhibit long range dependencies which can only be captured by computationally challenging, complex models. Sequence data arising from natural processes also often exhibits power-law properties, yet common sequence models do not capture such properties. The sequence memoizer is a new hierarchical Bayesian model for discrete sequence data that captures long range dependencies and power-law characteristics, while remaining computationally attractive. Its utility as a language model and general purpose lossless compressor is demonstrated. 1. intRoDuction It is an age-old quest to predict what comes next in sequences. Fortunes have been made and lost on the success and failure of such predictions. Heads or tails? Will the stock market go up by 5% tomorrow? Is the next card drawn from the deck going to be an ace? Does a particular sequence of nucleotides appear more often then usual in a DNA sequence? In a sentence, is the word that follows the

Journal

Communications of the ACMAssociation for Computing Machinery

Published: Feb 1, 2011

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off