Using artificial intelligence to control digital manufacturing
Researchers train a machine-learning model to monitor and adjust the 3D printing process to correct errors in real-time.
Researchers train a machine-learning model to monitor and adjust the 3D printing process to correct errors in real-time.
The technique opens a door to manufacturing of pressure-monitoring bandages, shade-shifting fabrics, or touch-sensing robots.
Evaluation Incubators to provide technical assistance, training, funding to help partners design randomized evaluations of housing stability strategies and state and local programs.
New stamp-sized ultrasound adhesives produce clear images of heart, lungs, and other internal organs.
To ensure a quick halt, brain circuit architecture avoids a slow process of integration in favor of quicker differentiation, study finds.
Using a randomized field experiment, researchers discover that Wikipedia articles affect judges’ legal reasoning.
The MIT Mobility Initiative welcomes five inaugural industry members to advance safe, clean, and inclusive mobility.
Cheap and quick to produce, these digitally manufactured plasma sensors could help scientists predict the weather or study climate change.
Neuroscience professor and Science Hub investigator Ted Adelson explains how simulating the sense of touch with a camera can make robots smarter.
Single-shot spectroscopy techniques provide researchers with a new understanding of a mysterious light-driven process.
“Interpretability methods” seek to shed light on how machine-learning models make predictions, but researchers say to proceed with caution.
Nerve cells regulate and routinely refresh the collection of calcium channels that enable them to send messages across circuit connections.
Researchers have found a material that can perform much better than silicon. The next step is finding practical and economic ways to make it.
The findings of a large-scale screen could help researchers design nanoparticles that target specific types of cancer.
Methods that make a machine-learning model’s predictions more accurate overall can reduce accuracy for underrepresented subgroups. A new approach can help.