Discrete Few-Shot Learning for Pan Privacy

TL;DR

This paper introduces a novel approach to few-shot image recognition using discrete embeddings and cryptographic protocols to enhance privacy, providing a baseline for future research in privacy-preserving recognition systems.

Contribution

It presents the first baseline results for discrete few-shot learning with a cryptographic protocol ensuring pan privacy in recognition systems.

Findings

First baseline results for discrete few-shot learning.

Cryptographic protocol guarantees pan privacy.

Practical implementation for privacy-preserving recognition.

Abstract

In this paper we present the first baseline results for the task of few-shot learning of discrete embedding vectors for image recognition. Few-shot learning is a highly researched task, commonly leveraged by recognition systems that are resource constrained to train on a small number of images per class. Few-shot systems typically store a continuous embedding vector of each class, posing a risk to privacy where system breaches or insider threats are a concern. Using discrete embedding vectors, we devise a simple cryptographic protocol, which uses one-way hash functions in order to build recognition systems that do not store their users' embedding vectors directly, thus providing the guarantee of computational pan privacy in a practical and wide-spread setting.