Search Stories

Researchers from MIT and elsewhere developed an algorithm that automatically and dynamically determines whether a machine learning to complete a task should try to mimic its teacher or explore on its own through trial-and-error. This algorithm enabled simulated student machines to learn tasks faster and more effectively than other techniques.

A more effective way to train machines for uncertain, real-world situations

Researchers develop an algorithm that decides when a “student” machine should follow its teacher, and when it should learn on its own.

May 31, 2023

Split screen with a purple image on the left that says CoCo, and on the right, a photo of a man and woman sitting with the MIT dome behind them. Both are wearing purple tinted glasses and flashing a hand sign.

CoCo: A real-time co-creative learning platform for young people

Media Lab PhD students Manuj Dhariwal and Shruti Dhariwal introduce a new paradigm of creative collaboration in digital learning environments.

May 30, 2023

Munther Dahleh speaks at a podium in an indoor space

Celebrating the impact of IDSS

A two-day conference at MIT reflected on the impact of the Institute for Data, Systems, and Society since its launch, as founding Director Munther Dahleh prepares to step down.

May 26, 2023

Cluster of pink blobs, depicting the Acinetobacter baumannii bacteria, are on a blue background with laser lines and glowing dots. Slightly left of center is a shooting scope that shows a clear view of the bacteria, while the rest of the background is blurry.

Using AI, scientists find a drug that could combat drug-resistant infections

The machine-learning algorithm identified a compound that kills Acinetobacter baumannii, a bacterium that lurks in many hospital settings.

May 25, 2023

Closeup image of a gauge with the word "SCORE" in bold. Levels are seen in rainbow colors with the options very bad, poor, fair, good, very good, and excellent. The needle is pointing to excellent.

Probabilistic AI that knows how well it’s working

It’s more important than ever for artificial intelligence to estimate how accurately it is explaining data.

May 25, 2023

Illustration of two humanoid robots standing in a kitchen, mixing fluids on a table to make a batter

Helping robots handle fluids

Researchers create a new simulation tool for robots to manipulate complex fluids in a step toward helping them more effortlessly assist with daily tasks.

May 24, 2023

2 rows of 3 images each. The images compare models of a horse and cow’s volume. The 3D models are sliced like clay, and the inside of the figures show a checkerboard of white and green or blue patterns. In the images comparing the volume of the two sliced figures, the checkerboards show the same number of squares, but the squares are stretched and pulled.

A better way to match 3D volumes

By mapping the volumes of objects, rather than their surfaces, a new technique could yield solutions to computer graphics problems in animation and CAD.

May 24, 2023

A photo shows a wooden table and wooden chairs that are glowing red, as if highlighted.

Researchers use AI to identify similar materials in images

This machine-learning method could assist with robotic scene understanding, image editing, or online recommendation systems.

May 23, 2023

Three MIT-led projects awarded MURI funding for 2023

Through the Multidisciplinary University Research Initiative, the US Department of Defense supports research projects in areas of critical importance to national defense.

May 18, 2023

Nine academics pose standing side by side while smiling for the camera in a room with a glass wall and the Boston skyline behind them

Is medicine ready for AI? Doctors, computer scientists, and policymakers are cautiously optimistic

With the artificial intelligence conversation now mainstream, the 2023 MIT-MGB AI Cures conference saw attendance double from previous years.

May 17, 2023

On left and right are portraits of Acemoglu and Johnson. In middle, is the cover of their book, which says in black and in golden embossed letters, “Our 1000-year Struggle over Technology & Prosperity; Power and Progress; Daron Acemoglu, co-author of Why Nations Fail; Simon Johnson, co-author of 13 Bankers.”

An AI challenge only humans can solve

In their new book, “Power and Progress,” Daron Acemoglu and Simon Johnson ask whether the benefits of AI will be shared widely or feed inequality.

May 17, 2023

13 portrait photos of MIT's Fulbright winners

Thirteen from MIT win 2023 Fulbright fellowships

The Fulbright US Student Program funds opportunities for research, graduate study, and teaching abroad.

May 15, 2023

Headshots of the five newly elected NAS members in a photo collage, with three in an upper row and two in a lower row

Five MIT faculty elected to the National Academy of Sciences for 2023

Joshua Angrist, Gang Chen, Catherine Drennan, Dina Katabi, Gregory Stephanopoulos, and seven additional alumni are recognized by their peers for their outstanding contributions to research.

May 11, 2023

Bryan Landman (left) and Ken Griffin sit on opposite sides of a small coffee table, conversing onstage at the Wong Auditorium. A backdrop with MIT logos is behind them.

Success at the intersection of technology and finance

Citadel founder and CEO Ken Griffin visits MIT, discusses how technology will continue to transform trading and investing.

May 10, 2023

A stylized rendering of a green judge’s gavel rests on a reflective surface. Red “thumbs down” icons are on top in the background, and green “thumbs up” reflections are on bottom of the image.

Study: AI models fail to reproduce human judgements about rule violations

Models trained using common data-collection techniques judge rule violations more harshly than humans would, researchers report.