Algorithms

In a study using machine-generated art to understand what makes a picture memorable, researchers found that the images people remembered best had bright colors, simple backgrounds, and subjects centered prominently in the frame.

What makes an image memorable? Ask a computer

An artificial intelligence model developed at MIT shows in striking detail what makes some images stick in our minds.

November 1, 2019

One modular robotic cube snaps into place with rest of the M-blocks.

Self-transforming robot blocks jump, spin, flip, and identify each other

Developed at MIT’s Computer Science and Artificial Intelligence Laboratory, robots can self-assemble to form various structures with applications including inspection.

October 30, 2019

MIT engineers have developed a system for autonomous vehicles that senses tiny changes in shadows on the ground to determine if there’s a moving object coming around the corner, such as when another car is approaching from behind a pillar in a parking garage.

Helping autonomous vehicles see around corners

By sensing tiny changes in shadows, a new system identifies approaching objects that may cause a collision.

October 27, 2019

Using a supercomputing system, MIT researchers developed a model that captures what global web traffic could look like on a given day, including previously unseen isolated links (left) that rarely connect but seem to impact core web traffic (right).

Supercomputer analyzes web traffic across entire internet

Modeling web traffic could aid cybersecurity, computing infrastructure design, Internet policy, and more.

October 27, 2019

MIT researchers have developed a trajectory-planning model that helps drones fly more safely at high speeds through previously unexplored areas, which could aid search-and-rescue missions through dense forests.

System prevents speedy drones from crashing in unfamiliar areas

Drones can fly at high speeds to a destination while keeping safe “backup” plans if things go awry.

October 25, 2019

A key to compiling the novel Omnipush dataset was building modular objects (pictured) that enabled the robotic system to capture a vast diversity of pushing behavior. The central pieces contain markers on their centers and points so a motion-detection system can detect their position within a millimeter.

Pushy robots learn the fundamentals of object manipulation

Systems “learn” from novel dataset that captures how pushed objects move, to improve their physical interactions with new objects.

October 21, 2019

At left, EdgeConv, a method developed at MIT, successfully finds meaningful parts of 3D shapes, like the surface of a table, wings of an airplane, and wheels of a skateboard. At right is the ground-truth comparison.

Deep learning with point clouds

Research aims to make it easier for self-driving cars, robotics, and other applications to understand the 3D world.

October 21, 2019

MIT Professor Regina Barzilay (left) and CSAIL PhD student Tal Schuster are studying detectors of machine-generated text.

Better fact-checking for fake news

New research reveals biases in fake news datasets and improves the use of automatic detectors.

October 17, 2019

A new algorithm speeds up the planning process for robotic grippers to manipulate objects using the surrounding environment.

Giving robots a faster grasp

An algorithm speeds up the planning process robots use to adjust their grip on objects, for picking and sorting, or tool use.

October 17, 2019

A novel model developed at MIT recovers valuable data lost from images and video that have been “collapsed” into lower dimensions. It can, for instance, recreate video from motion-blurred images or from cameras that capture people’s movement around corners as vague one-dimensional lines.

Recovering “lost dimensions” of images and video

Model could recreate video from motion-blurred images and “corner cameras,” may someday retrieve 3D data from 2D medical images.

October 16, 2019

A new technique for training video recognition models is up to three times faster than current state-of-the-art methods while improving runtime performance on mobile devices. The work was recently highlighted by Dario Gil (above), director of IBM Research, at the MIT-IBM Watson AI Lab’s AI Research Week in Cambridge, Massachusetts.

Faster video recognition for the smartphone era

MIT and IBM researchers offer a new method to train and run deep learning models more efficiently.

October 11, 2019

A simulation of early galaxy formation under three dark matter scenarios. In a universe filled with cold dark matter, early galaxies would first form in bright halos (far left). If dark matter is instead warm, galaxies would form first in long, tail-like filaments (center). Fuzzy dark matter would produce similar filaments, though striated (far right), like the strings of a harp.

This is how a “fuzzy” universe may have looked

Scientists simulate early galaxy formation in a universe of dark matter that is ultralight, or “fuzzy,” rather than cold or warm.

October 3, 2019

A system designed by researchers at MIT and elsewhere enables interconnected smart devices to cooperatively pinpoint their positions in noisy environments where GPS usually fails, which is useful for emerging “localization-of-things” applications.

System helps smart devices find their position

Connected devices can now share position information, even in noisy, GPS-denied areas.

October 2, 2019

An algorithm developed at MIT takes MRI images of placentas (top) and flattens them to be more easily analyzed (middle and bottom).

Using algorithms to build a map of the placenta

New technique stretches out MRI scans of placentas so they can be more accurately analyzed, and shows the potential of MRI for pregnancy monitoring.

October 2, 2019

Trevor Henderson in the record library at WMBR, MIT’s student radio station.

Using math to blend musical notes seamlessly

Algorithm enables one audio signal to glide into another, recreating the “portamento” effect of some musical instruments.

September 27, 2019

MIT News | Massachusetts Institute of Technology - On Campus and Around the world

Browse By

Topics

Departments

Centers, Labs, & Programs

Schools

Topic