Tue Mar 07 2023
Sun Mar 05 2023

Towards Democratizing Joint-Embedding Self-Supervised Learning

Self-supervised learning
Machine Learning
Computer Vision
Improve the performance of JE-SSL
Easily evaluate JE-SSL methods
Train well-known SSL methods like SimCLR

Joint Embedding Self-Supervised Learning (JE-SSL) has seen rapid developments in recent years, due to its promise to effectively leverage large unlabeled data. The development of JE-SSL methods was driven primarily by the search for ever increasing downstream classification accuracies, using huge computational resources, and typically built upon insights and intuitions inherited from a close parent JE-SSL method. In this work, we debunk several such ill-formed a priori ideas in the hope to unleash the full potential of JE-SSL free of unnecessary limitations.

Provides actionable insights to improve Joint Embedding Self-Supervised Learning (JE-SSL) by debunking misconceptions and introducing an optimized PyTorch library for SSL.

Alexa Arena: A User-Centric Interactive Platform for Embodied AI

Embodied AI
Artificial Intelligence
Robotics
Advance research in Human Robot Interaction (HRI)
Develop embodied agents for robotic task completion challenges
Create gamified robotic tasks for general human users

We introduce Alexa Arena, a user-centric simulation platform for Embodied AI (EAI) research. Alexa Arena provides a variety of multi-room layouts and interactable objects, for the creation of human-robot interaction (HRI) missions. With user-friendly graphics and control mechanisms, Alexa Arena supports the development of gamified robotic tasks readily accessible to general human users, thus opening a new venue for high-efficiency HRI data collection and EAI system evaluation.

Provides a platform for developing embodied agents for robotic task completion challenges, with a focus on advancing research in Human Robot Interaction (HRI).

Thu Mar 02 2023
Wed Mar 01 2023
Tue Feb 28 2023
Mon Feb 27 2023