甲骨文公司
计算机科学
管道(软件)
人工智能
监督学习
相似性(几何)
机器学习
半监督学习
先验与后验
主动学习(机器学习)
简单(哲学)
语义相似性
情报检索
图像(数学)
人工神经网络
软件工程
认识论
哲学
程序设计语言
作者
Vivien Cabannes,Léon Bottou,Yann LeCun,Randall Balestriero
标识
DOI:10.1109/iccv51070.2023.01491
摘要
Self-Supervised Learning (SSL) has emerged as the solution of choice to learn transferable representations from unlabeled data. However, SSL requires to build samples that are known to be semantically akin, i.e. positive views. Requiring such knowledge is the main limitation of SSL and is often tackled by ad-hoc strategies e.g. applying known data-augmentations to the same input. In this work, we formalize and generalize this principle through Positive Active Learning (PAL) where an oracle queries semantic relationships between samples. PAL achieves three main objectives. First, it unveils a theoretically grounded learning framework beyond SSL, based on similarity graphs, that can be extended to tackle supervised and semi-supervised learning depending on the employed oracle. Second, it provides a consistent algorithm to embed a priori knowledge, e.g. some observed labels, into any SSL losses without any change in the training pipeline. Third, it provides a proper active learning framework yielding low-cost solutions to annotate datasets, arguably bringing the gap between theory and practice of active learning that is based on simple-to-answer-by-non-experts queries of semantic relationships between inputs.
科研通智能强力驱动
Strongly Powered by AbleSci AI