The Lovász Hinge: A Novel Convex Surrogate for Submodular Losses
This provides a tractable solution for set prediction tasks, such as on PASCAL VOC and COCO datasets, addressing a bottleneck in machine learning for structured prediction.
The paper tackles the problem of learning with non-modular losses in set prediction by proposing the Lovász hinge, a novel convex surrogate for submodular losses, which achieves O(p log p) complexity for gradient computation, improving over NP-hard methods.
Learning with non-modular losses is an important problem when sets of predictions are made simultaneously. The main tools for constructing convex surrogate loss functions for set prediction are margin rescaling and slack rescaling. In this work, we show that these strategies lead to tight convex surrogates iff the underlying loss function is increasing in the number of incorrect predictions. However, gradient or cutting-plane computation for these functions is NP-hard for non-supermodular loss functions. We propose instead a novel surrogate loss function for submodular losses, the Lovász hinge, which leads to O(p log p) complexity with O(p) oracle accesses to the loss function to compute a gradient or cutting-plane. We prove that the Lovász hinge is convex and yields an extension. As a result, we have developed the first tractable convex surrogates in the literature for submodular losses. We demonstrate the utility of this novel convex surrogate through several set prediction tasks, including on the PASCAL VOC and Microsoft COCO datasets.