New algorithms enable efficient machine learning with ... - MIT News
Jul 30, 2025 · MIT researchers designed a computationally efficient algorithm for machine learning with symmetric data that also requires fewer data for training than conventional …
Machine learning | MIT News | Massachusetts Institute of …
4 days ago · MIT CSAIL and LIDS researchers developed a mathematically grounded system that lets soft robots deform, adapt, and interact with people and objects, without violating safety limits.
AI learns how vision and sound are connected, without human
May 22, 2025 · A new machine-learning model learns to pinpoint exactly where a particular sound occurs in a video clip without the need for human intervention. The model could have …
A faster way to solve complex planning problems - MIT News
Apr 16, 2025 · MIT researchers developed a machine-learning-guided technique to solve complex, long-horizon planning problems more efficiently than some traditional approaches, …
An anomaly detection framework anyone can use - MIT News
May 28, 2025 · MIT PhD candidate Sarah Alnegheimish’s research interests reside at the intersection of machine learning and systems engineering. Her objective: to make machine …
“Periodic table of machine learning” could fuel AI discovery
Apr 23, 2025 · After uncovering a unifying algorithm that links more than 20 common machine-learning approaches, MIT researchers organized them into a “periodic table of machine …
New model predicts a chemical reaction’s point of no return
Apr 23, 2025 · MIT researchers developed a machine-learning model that can predict the structures of transition states of chemical reactions in less than a second, with high accuracy. …
New machine-learning application to help researchers predict
Jul 24, 2025 · MIT researchers developed ChemXploreML, a new desktop application that provides an automated and user-friendly approach to predicting molecular properties using …
MIT News - Machine learning
The approach combines physics and machine learning to avoid damaging disruptions when powering down tokamak fusion machines. New prediction model could improve the reliability …
Learning the language of molecules to predict their properties
Jul 7, 2023 · A new MIT framework uses machine learning to simultaneously predict molecular properties and generate new molecules using only a small amount of data for training.