Scalable optimization of multivariate performance measures in multi-instance multi-label learning

Abstract

The problem of multi-instance multi-label learning (MIML) requires a bag of instances to be assigned a set of labels most relevant to the bag as a whole. The problem finds numerous applications in machine learning, computer vision, and natural language processing settings where only partial or distant supervision is available. We present a novel method for optimizing multivariate performance measures in the MIML setting. Our approach MIMLperf uses a novel plug-in technique and offers a seamless way to optimize a vast variety of performance measures such as macro and micro-F measure, average precision, which are performance measures of choice in multi-label learning domains. MIMLperf offers two key benefits over the state of the art. Firstly, across a diverse range of benchmark tasks, ranging from relation extraction to text categorization and scene classification, MIMLperf offers superior performance as compared to state of the art methods designed specifically for these tasks. Secondly, MIMLperf operates with significantly reduced running times as compared to other methods, often by an order of magnitude or more.