/pal

Active self-supervised learning with pairwise comparison

Primary LanguageJupyter NotebookMIT LicenseMIT

Experiments for active self-supervised learning

This repository contains the code to reproduce the experiments in the paper: "Active Self-Supervised Learning: A Few Low-Cost Relationships Are All You Need" by Vivien Cabannes, L'eon Bottou, Yann LeCun and Randall Balestriero (ICCV, 2023).

Installation

The code is written in Python 3.6. It depends on the library https://github.com/VivienCabannes/rates.

License

Those notebooks are MIT licensed, as precised in the LICENSE file.