Learning to Remember Rare Events

TL;DR

This paper introduces a scalable, differentiable memory module for deep neural networks that enables effective life-long and one-shot learning of rare events across various architectures and tasks.

Contribution

It presents a large-scale, end-to-end trainable memory module that enhances neural networks' ability to remember rare events over long periods without resets.

Findings

Achieved state-of-the-art one-shot learning on Omniglot dataset.

Enabled life-long one-shot learning in recurrent neural networks for machine translation.

Demonstrated versatility by integrating with different neural network architectures.

Abstract

Despite recent advances, memory-augmented deep neural networks are still limited when it comes to life-long and one-shot learning, especially in remembering rare events. We present a large-scale life-long memory module for use in deep learning. The module exploits fast nearest-neighbor algorithms for efficiency and thus scales to large memory sizes. Except for the nearest-neighbor query, the module is fully differentiable and trained end-to-end with no extra supervision. It operates in a life-long manner, i.e., without the need to reset it during training. Our memory module can be easily added to any part of a supervised neural network. To show its versatility we add it to a number of networks, from simple convolutional ones tested on image classification to deep sequence-to-sequence and recurrent-convolutional models. In all cases, the enhanced network gains the ability to remember and do life-long one-shot learning. Our module remembers training examples shown many thousands of steps in the past and it can successfully generalize from them. We set new state-of-the-art for one-shot learning on the Omniglot dataset and demonstrate, for the first time, life-long one-shot learning in recurrent neural networks on a large-scale machine translation task.