Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions
We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples o...
see more
Find all the Top AIManufacturing papers. Links to pdf, code repos and demos are provided.
We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples o...
see more
We propose a low cost and effective way to combine a free simulation software
and free CAD models for modeling human-object interaction in order to im...
see more
Presents PatchCore, a novel algorithm designed for cold-start anomaly detection in industrial settings, where only non-defective images are available ...
see more
Introduces PIVOT, an iterative visual prompting methodology that enables vision-language models (VLMs) to produce actionable outputs for spatial tasks...
see more
Explores the potential of low-cost robotic systems to perform fine manipulation tasks through imitation learning. It introduces a novel algorithm, Act...
see more
Addresses the limitations of current robotic grasping techniques that rely solely on visual input by introducing a method that incorporates tactile fe...
see more
Explores the application of deep reinforcement learning (RL) for performing complex industrial insertion tasks, specifically in environments with visu...
see more
Introduces a novel model-based reinforcement learning algorithm designed for robotic manipulation tasks that significantly reduces the amount of exper...
see more
Presents QT-Opt, a scalable deep reinforcement learning framework designed for learning vision-based dynamic manipulation skills, specifically focusin...
see more
Addresses the challenge of viewpoint-invariant visual servoing in robotic manipulation, proposing an innovative deep recurrent controller that learns ...
see more
Presents MT-Opt, a novel approach to continuous multi-task robotic reinforcement learning that allows robotic systems to learn a diverse set of skills...
see more
Explores the application of deep reinforcement learning (DRL) in robotic manipulation, specifically focusing on improving training efficiency for real...
see more