There has been a recent trend in training neural networks to replace data structures that have been crafted by hand, with an aim for faster execution, better accuracy, or greater compression. In this setting, a neural data structure is instantiated by training a network over many epochs of its inputs until convergence. In applications where inputs arrive at high throughput, or are ephemeral, training a network from scratch is not practical. This motivates the need for few-shot neural data structures. In this paper we explore the learning of approximate set membership over a set of data in one-shot via meta-learning. We propose a novel memory architecture, the Neural Bloom Filter, which is able to achieve significant compression gains over classical Bloom Filters and existing memory-augmented neural networks.
@article{arxiv.1906.04304,
title = {Meta-Learning Neural Bloom Filters},
author = {Jack W Rae and Sergey Bartunov and Timothy P Lillicrap},
journal= {arXiv preprint arXiv:1906.04304},
year = {2019}
}