DeepAI AI Chat
Log In Sign Up

Leveraging Hidden Structure in Self-Supervised Learning

06/30/2021
by   Emanuele Sansone, et al.
0

This work considers the problem of learning structured representations from raw images using self-supervised learning. We propose a principled framework based on a mutual information objective, which integrates self-supervised and structure learning. Furthermore, we devise a post-hoc procedure to interpret the meaning of the learnt representations. Preliminary experiments on CIFAR-10 show that the proposed framework achieves higher generalization performance in downstream classification tasks and provides more interpretable representations compared to the ones learnt through traditional self-supervised learning.

READ FULL TEXT

page 4

page 8

11/15/2022

Homomorphic Self-Supervised Learning

In this work, we observe that many existing self-supervised learning alg...
04/01/2022

Simplicial Embeddings in Self-Supervised Learning and Downstream Classification

We introduce Simplicial Embeddings (SEMs) as a way to constrain the enco...
03/12/2021

Information Maximization Clustering via Multi-View Self-Labelling

Image clustering is a particularly challenging computer vision task, whi...
09/16/2022

Self-Supervised Learning with an Information Maximization Criterion

Self-supervised learning allows AI systems to learn effective representa...
11/19/2020

Robot Gaining Accurate Pouring Skills through Self-Supervised Learning and Generalization

Pouring is one of the most commonly executed tasks in humans' daily live...
04/19/2022

Rumor Detection with Self-supervised Learning on Texts and Social Graph

Rumor detection has become an emerging and active research field in rece...
06/04/2019

Information Competing Process for Learning Diversified Representations

Learning representations with diversified information remains an open pr...