AI Briefing#imitation-learning
The latest AI and developer news about #imitation-learning, with the original source and a short summary.
The latest AI and developer news about #imitation-learning, with the original source and a short summary.
The new MA-BC method pools demonstrations only when expert actions agree, providing upper and lower bounds on sample complexity.
OpenArm has released an open-source humanoid arm for teleoperation and imitation learning.
Salesforce researchers proposed an On-Policy Correction pipeline to overcome the limitations of imitation learning for weaker models and facilitate the co-evolution of harnesses and models.
We propose MoMo, a framework that allows robots to adjust their motion style according to the task environment through spatiotemporal action tokenization.
Xiaomi has unveiled Xiaomi-Robotics-1, a robot policy model built on 100,000 hours of embodiment-agnostic pretraining.
The research proposed a new imitation learning methodology that leverages expert demonstrations containing only state information to enable stable learning even in offline settings.
We propose a new algorithm that improves the stability of offline Learning from Observation (LfO), which imitates expert behavior using only state information.
Sim-to-Real and Real-to-Sim technologies that narrow the gap between simulation and reality are the core drivers of Physical AI advancement.
A DAgger pipeline that trains a tactical AI on human demonstrations in Ready or Not.