Cardinality-Aware Set Prediction and Top-$k$ Classification
This work addresses the challenge of improving top-k classification for machine learning applications by reducing unnecessary predictions, though it appears incremental as it builds on existing classification frameworks.
The paper tackles the problem of top-k classification by developing cardinality-aware set predictors that balance classification accuracy with low cardinality, introducing new loss functions and algorithms with theoretical guarantees. Experimental results on datasets like CIFAR-10, CIFAR-100, ImageNet, and SVHN demonstrate the effectiveness of their approach.
We present a detailed study of cardinality-aware top-$k$ classification, a novel approach that aims to learn an accurate top-$k$ set predictor while maintaining a low cardinality. We introduce a new target loss function tailored to this setting that accounts for both the classification error and the cardinality of the set predicted. To optimize this loss function, we propose two families of surrogate losses: cost-sensitive comp-sum losses and cost-sensitive constrained losses. Minimizing these loss functions leads to new cardinality-aware algorithms that we describe in detail in the case of both top-$k$ and threshold-based classifiers. We establish $H$-consistency bounds for our cardinality-aware surrogate loss functions, thereby providing a strong theoretical foundation for our algorithms. We report the results of extensive experiments on CIFAR-10, CIFAR-100, ImageNet, and SVHN datasets demonstrating the effectiveness and benefits of our cardinality-aware algorithms.