Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Unifying logical and statistical AI with Markov logic
Domingos P., Lowd D. Communications of the ACM62 (7):74-83,2019.Type:Article
Date Reviewed: Sep 20 2019

A Markov logic network (MLN) is a probabilistic logic that endows the first-order logic with a degree of uncertainty by applying Markov networks. It represents an active area of research, introduced in the early 2000s by the very authors of the present article. It is already supported by a body of published results, and successfully put into practice in domains like chemistry, medicine, and social relations. Thus, MLN has already proved its strength in predicting protein interaction and functions, predicting work relationships, knowledge base tuning, and handling medical knowledge bases.

The article provides an overview of this theory and highlights its up-to-date achievements, from both theoretical and practical perspectives. At the outset, it reviews recent statistical artificial intelligence (AI) approaches, including how they relate to Markov logic and their ability to cope with uncertainty issues. Subsequently, the article presents the chain of elements that makes the transition from first-order logic to Markov logic and MLN. An important part of the work discusses key problems related to MLN, such as handling inference and learning and configuration issues, taking into account efficiency, computational cost, and the ability to accommodate new concepts. The authors specify some extensions and remodeling applied to Markov logic to match special requirements. Regarding more practical issues, the article reviews the classes of tasks where MLN has been applied so far. A discussion of some available open-source packages that implement various Markov logic algorithms is also valuable.

The ideas presented should encourage readers to more deeply explore this new side of AI, and possibly expand its application domain.

Reviewer:  Svetlana Segarceanu Review #: CR146699 (1912-0452)
Bookmark and Share
 
General (I.2.0 )
 
 
Markov Processes (G.3 ... )
 
Would you recommend this review?
yes
no
Other reviews under "General": Date
Artificial experts: social knowledge and intelligent machines
Collins H., MIT Press, Cambridge, MA, 1990. Type: Book (9780262031684)
Apr 1 1991
Catalogue of artificial intelligence techniques
Bundy A., Springer-Verlag New York, Inc., New York, NY, 1990. Type: Book (9780387529592)
Aug 1 1991
Knowledge and inference
Nagao M., Academic Press Prof., Inc., San Diego, CA, 1990. Type: Book (9780125136624)
Oct 1 1991
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy