Maximum Entropy Modeling (Part-1)

Một mô hình quan trọng trong NLP mà mình chưa nắm được một cách tường tận.

Bài này tập hợp các good link để bắt đầu với mô hình này:


Link từ trang Web của Zhang Le. Tổng hợp hầu như tất cả các thứ liên quan đến Maxent từ các bài báo, công cụ đến ứng dụng, etc.

Xin phép tác giả được đăng lại ở trang Blog của mình 🙂

Maximum Entropy Modeling

This page dedicates to a general-purpose machine learning technique called Maximum Entropy Modeling (MaxEnt for short). On this page you will find:

What is Maximum Entropy Modeling

In his famous 1957 paper, Ed. T. Jaynes wrote:
Information theory provides a constructive criterion for setting up probability distributions on the basis of partial knowledge, and leads to a type of statistical inference which is called the maximum entropy estimate. It is least biased estimate possible on the given information; i.e., it is maximally noncommittal with regard to missing information.
That is to say, when characterizing some unknown events with a statistical model, we should always choose the one that has Maximum Entropy.

Maximum Entropy Modeling has been successfully applied to Computer Vision, Spatial Physics, Natural Language Processing and many other fields. This page will focus on applying Maxent to Natural Language Processing (NLP).

The concept of Maximum Entropy can be traced back along multiple threads to Biblical times. However, not until the late of 21st century has computer become powerful enough to handle complex problems with statistical modeling technique like Maxent.

Maximum Entropy was first introduced to NLP area by Berger, et al (1996) and Della Pietra, et al. 1997. Since then, Maximum Entropy technique (and the more general framework Random Fields) has enjoyed intensive research in NLP community.

Tutorials for Maximum Entropy Modeling

Here is an (incomplete) list of tutorials & introduction for Maximum Entropy Modeling.

Maxent related software

Here is an incomplete list of software found on the net that are related to Maximum Entropy Modeling.

Annotated papers on Maximum Entropy Modeling in NLP

Here is a list of recommended papers on Maximum Entropy Modeling with brief annotation.

  • A Maximum Entropy Approach to Natural Language Processing (Berger, et al. 1996)
    A must read paper on applying maxent technique to Natural Language Processing. This paper describes maxent in detail and presents an Increment Feature Selection algorithm for increasingly construct a maxent model as well as several examples in statistical Machine Translation.
  • Inducing Features of Random Fields (Della Pietra, et al. 1997)
    Another must read paper on maxent. It deals with a more general frame work: Random Fields and proposes an Improved Iterative Scaling algorithm for estimating parameters of Random Fields. This paper gives theoretical background to Random Fields (and hence Maxent model). A greedy Field Induction method was presented to automatically construct a detail random fields from a set of atomic features. An word morphology application for English was developed. longer version.
  • Adaptive Statistical Language Modeling: A Maximum Entropy Approach (Rosenfeld, 1994)
    This paper applies ME technique to statistical language modeling task. More specifically, it builds a conditional Maximum Entropy model that incorporates traditional N-gram, distant N-gram and trigger pair features. Significantly perplexity reduction over baseline trigram model was reported. Later, Rosenfeld and his group proposed a Whole Sentence Exponential Model that overcome the computation bottleneck of conditional ME model. You can find more on my SLM page.
  • Maximum Entropy Models For Natural Language Ambiguity Resolution (Ratnaparkhi, 1998)
    This dissertation discusses the application of maxent model to various Natural Language Dis-ambiguity tasks in detail. Several problems were attacked within the ME framework: sentence boundary detection, part-of-speech tagging, shallow parsing and text categorization. Comparison with other machine learning technique (Naive Bayes, Transform Based Learning, Decision Tree etc.) was given. Ratnaparkhi also had a short introduction paper on ME.
  • The Improved Iterative Scaling Algorithm: A Gentle Introduction
    This paper describes IIS algorithm in detail. The description is easier to understand than (Della Pietra, et al. 1997), which involves more mathematical notations.
  • Stochastic Attribute-Value Grammars (Abney, 1997)
    Abney applies Improved Iterative Scaling algorithm to parameters estimation of Attribute-Value grammars, which can not be corrected calculated by ERF method (though it works on PCFG). Random Fields is the model of choice here with a general Metropolis-Hasting Sampling on calculating feature expectation under newly constructed model.
  • A comparison of algorithms for maximum entropy parameter estimation (Malouf, 2003)
    Four iterative parameter estimation algorithms are compared on several NLP tasks. L-BFGS is observed to be the most effective parameter estimation method for Maximum Entropy model, much better than IIS and GIS. (Wallach 02) reported similar results on parameter estimation of Conditional Random Fields. Here is Malouf’s Maximum Entropy Parameter Estimation software.
  • A Mathematical Theory of Communication
    Claude Elwood Shannon‘s influential 1948 paper that laid the foundation of information theory and changed the whole world since then. I see no reason who has read the above papers does not want to read this one.
  • Information Theory and Statistical Mechanics (Jaynes, E. T., 1957)
    Having read all the above papers? Well, it’s time to have a look at this one. Edwin Thompson Jaynes presented some insightful results of maximum entropy principle in this 1957 paper published in Physics Reviews. This is also his first paper in information theory. Interestingly, this influential work was published over the objection of a reviewer.

Other recommended papers:

Other MaxEnt related resources on the web

Từ trang Web của Berger

Logistic Regression

*********** Toolkit for Maxent

[1] Tsuruoka’s Maxent:

[2] Mallet:


4 Responses

  1. Há há, search phát ra cái này đang tưởng của ai, hoá ra là đồng chí Minh nhà mình :D. Good job!

  2. @Anh Cường: Em lập ra để note những gì search được cho đỡ quên thôi chứ có gì hay ho lắm đâu anh 😀

  3. Hi LightRiver…

    nice post, friend…
    now, i have more references for my final project, about NLP too
    thank u very much 😀

  4. […] Tại sao MaxEnt được ứng dụng trong nlp? 2. Ý nghĩa của F score là gì? khi không sử dụng Accuracy. […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: