Maximizing ‘bang per bit’ to achieve state of the art results

(Submitted on 2 Nov 2019 (v1), last revised 5 Nov 2019 (this version, v2))

Abstract: Building on recent work on capsule networks, we propose a new, general-purpose form of ``routing by agreement'' that activates output capsules in a layer as a function of their net benefit to use and net cost to ignore input capsules from earlier layers. To illustrate the usefulness of our routing algorithm, we present two capsule networks that apply it in different domains: vision and language. The first network achieves new state-of-the-art accuracy of 99.1% on the smallNORB visual recognition task with fewer parameters and an order of magnitude less training than previous capsule models, and we find evidence that it learns to perform a form of ``reverse graphics.'' The second network achieves new state-of-the-art accuracies on the root sentences of the Stanford Sentiment Treebank: 58.5% on fine-grained and 95.6% on binary labels with a single-task model that routes frozen embeddings from a pretrained transformer as capsules. In both domains, we train with the same regime. Code is available at this https URL along with replication instructions.