A simpler method for learning to control a robot
Researchers develop a machine-learning technique that can efficiently learn to control a robot, leading to better performance with fewer data.
School of Engineering
A faster way to teach a robot
A new technique helps a nontechnical user understand why a robot failed, and then fine-tune it with minimal effort to perform a task effectively.
Armando Solar-Lezama named inaugural Distinguished College of Computing Professor
EECS professor appointed to new professorship in the MIT Schwarzman College of Computing.
AI helps household robots cut planning time in half
PIGINet leverages machine learning to streamline and enhance household robots’ task and motion planning, by assessing and filtering feasible solutions in complex environments.
AI helps household robots cut planning time in half
PIGINet leverages machine learning to streamline and enhance household robots’ task and motion planning, by assessing and filtering feasible solutions in complex environments.
A new way to look at data privacy
Researchers create a privacy technique that protects sensitive data while maintaining a machine-learning model’s performance.
Generative AI imagines new protein structures
MIT researchers develop “FrameDiff,” a computational tool that uses generative AI to craft new protein structures, with the aim of accelerating drug development and improving gene therapy.
3 Questions: Honing robot perception and mapping
Luca Carlone and Jonathan How of MIT LIDS discuss how future robots might perceive and interact with their environment.
Learning the language of molecules to predict their properties
This AI system only needs a small amount of data to predict molecular properties, which could speed up drug discovery and material development.
MIT scientists build a system that can generate AI models for biology research
BioAutoMATED, an open-source, automated machine-learning platform, aims to help democratize artificial intelligence for research labs.