Markov Logic: Theory, Algorithms and Applications

  • Parag Singla | University of Washington

AI systems must be able to reason about complex objects as well explicitly handle uncertainty. First order logic gives the formalism to handle the first. Probability gives the power to handle the latter. Combining the two has been a long standing goal of AI research. In this talk, I will present Markov Logic (Richardson & Domingos 06), which combines the power of full first order logic and Markov networks. Markov logic represents the underlying world by attaching real valued weights to formulas in first order logic. The formulas in Markov logic can be seen as defining templates for ground Markov networks. Carrying out propositional inference techniques in such models leads to explosion in time and memory. To overcome these problems, I will present the first algorithm for lifted probabilistic inference with results on real data: lifted belief propagation. Learning of the parameters (formula weights) is done using a voted perceptron algorithm. I will then present applications to the problems of entity resolution and identification of social relationships in consumer photo collections. I will conclude the talk with directions for future work.

Speaker Details

Parag received his bachelor’s degree in Computer Science & Engineering from IIT Bombay, India in 2002. He then joined University of Washington for graduate studies in Computer Science. He finished his Masters in 2004 and successfully defended his thesis in October 2008. His PhD advisor is Pedro Domingos. He has several publications in premiere AI, data mining and IR conferences such as AAAI, UAI, ICDM, PKDD and WWW and was the winner of the best paper award at PKDD-05. Parag has been a reviewer for the journals TKDD and AMAI and has also filed two patents. He has interned at top research institutions including IBM India Research Lab, Kodak Research Labs Rochester and Microsoft Research Redmond.Parag’s research areas include machine learning, data mining and social networks analysis. Specifically, his PhD research has been around the development of Markov logic, one of the most powerful models for statistical relational AI. He is also a co-author for Alchemy, an open source software for statistical relational AI developed at the University of Washington.Full Vita at: http://www.cs.washington.edu/homes/parag/parag-resume.pdf

Watch Next