Determinantal point processes for machine learning. (English) Zbl 1278.68240

Kulesza, Alex; Taskar, Ben

Found. Trends Mach. Learn. 5, No. 2-3, 123-286 (2012).

Summary: Determinantal point processes (DPPs) are elegant probabilistic models of repulsion that arise in quantum physics and random matrix theory. In contrast to traditional structured models like Markov random fields, which become intractable and hard to approximate in the presence of negative correlations, DPPs offer efficient and exact algorithms for sampling, marginalization, conditioning, and other inference tasks. We provide a gentle introduction to DPPs, focusing on the intuitions, algorithms, and extensions that are most relevant to the machine learning community, and show how DPPs can be applied to real-world applications like finding diverse sets of high-quality search results, building informative summaries by selecting diverse sentences from documents, modeling nonoverlapping human poses in images or video, and automatically building timelines of important news stories.

MSC:

68T05	Learning and adaptive systems in artificial intelligence
60G55	Point processes (e.g., Poisson, Cox, Hawkes processes)
68-02	Research exposition (monographs, survey articles) pertaining to computer science

Keywords:

machine learning; determinantal point processes; DPPs; Markov random fields

Software:

TOMS659