Exact and efficient top-K inference for multi-target prediction by querying separable linear relational models
This work addresses efficiency challenges in multi-target prediction for applications like collaborative filtering and biological network inference, though it is incremental as it builds on existing information retrieval methods.
The paper tackles the problem of efficiently computing top-K predictions in multi-target prediction tasks with large target spaces, such as collaborative filtering and multi-label classification, by proposing modifications of information retrieval methods for separable linear relational models. The results show that the threshold algorithm is highly scalable, often performing many orders of magnitude more efficiently than naive approaches.
Many complex multi-target prediction problems that concern large target spaces are characterised by a need for efficient prediction strategies that avoid the computation of predictions for all targets explicitly. Examples of such problems emerge in several subfields of machine learning, such as collaborative filtering, multi-label classification, dyadic prediction and biological network inference. In this article we analyse efficient and exact algorithms for computing the top-$K$ predictions in the above problem settings, using a general class of models that we refer to as separable linear relational models. We show how to use those inference algorithms, which are modifications of well-known information retrieval methods, in a variety of machine learning settings. Furthermore, we study the possibility of scoring items incompletely, while still retaining an exact top-K retrieval. Experimental results in several application domains reveal that the so-called threshold algorithm is very scalable, performing often many orders of magnitude more efficiently than the naive approach.