Machine learning

Researchers developed a simple application for devices that use Apple’s iOS operating system. The application flashes a small dot somewhere on the device’s screen, attracting the user’s attention, then briefly replaces it with either an “R” or an “L,” instructing the user to swipe either the right or left side of the screen. Correctly executing the swipe ensures that the user has act...

Eye-tracking system uses ordinary cellphone camera

Crowd-sourced data yields system that determines where mobile-device users are looking.

June 15, 2016

Artificial intelligence produces realistic sounds that fool humans

Video-trained system from MIT’s Computer Science and Artificial Intelligence Lab could help robots understand how objects interact with the world.

June 13, 2016

IDSS conversations: Guy Bresler

Assistant professor of electrical engineering and computer science discusses his work focusing on learning graphical models from data.

March 11, 2016

“One of the most intriguing aspects of this research is that we’ve found that there are indeed universal properties of correct code that you can learn from one set of applications and apply to another set of applications,” Martin Rinard says.

Recognizing correct code

Automatic bug-repair system fixes 10 times as many errors as its predecessors.

January 29, 2016

MIT researchers fed their physics prediction system videos of collisions, shown as screenshots in (a) and (b), which were then converted into simulations generated by the 3-D physics engine, shown in (c) and (d).

Computer model matches humans at predicting how objects move

“3-D physics engine” from the Computer Science and Artificial Intelligence Laboratory simulates the human brain to infer physical properties.

January 4, 2016

Researchers have developed a computational model of visual representation intended to reflect the way the brain works.

Machines that learn like people

Algorithms could learn to recognize objects from a few examples, not millions; may better model human cognition.

December 23, 2015

For each image, the MemNet algorithm creates a heat map identifying its most memorable and forgettable regions.

Deep-learning algorithm predicts photos’ memorability at “near-human” levels

Future versions of an algorithm from the Computer Science and Artificial Intelligence Lab could help with teaching, marketing, and memory improvement.

December 16, 2015

Humans and machines were given an image of a novel character (top) and asked to produce new versions. A machine generated the nine-character grid on the left.

Computer system passes “visual Turing test”

Character-drawing machines can fool human judges.

December 10, 2015

Predicting change in the Alzheimer’s brain

Combining MRI and other data helps machine-learning systems predict effects of neurodegenerative disease.

October 6, 2015

Flickr users tagged a photograph similar to this one
“architecture," “tourism," and "travel.” A machine-learning system that
used a novel training strategy developed at MIT proposed “sky," “roof,”
and “building”; when it used a conventional training strategy, it came up
with “art," “sky,” and "beach.”

More-flexible machine learning

Giving machine-learning systems “partial credit” during training improves image classification.

October 1, 2015

Learning language by playing games

System learns to play text-based computer game using only linguistic information.

September 23, 2015

Illustration of hands on a computer screen

More personalized online shopping

Startup’s platform analyzes data from multiple sources to better predict buying preferences.

September 15, 2015

The first layers (1 and 2) of a neural network trained to classify scenes seem to be tuned to geometric patterns of increasing complexity, but the higher layers (3 and 4) appear to be picking out particular classes of objects.

Object recognition for free

System designed to label visual scenes according to type turns out to detect particular objects, too.

May 8, 2015

Two-dimensional images of human faces (top row) and front views of three-dimensional models of the same faces, produced by both a new MIT system (middle row) and one of its predecessors (bottom row).

Graphics in reverse

Probabilistic programming does in 50 lines of code what used to take thousands.

April 13, 2015

Better machine learning

Kalyan Veeramachaneni tackles some of the biggest bottlenecks holding back the data science industry.

February 24, 2015

MIT News | Massachusetts Institute of Technology

Browse By

Topics

Departments

Centers, Labs, & Programs

Schools

Topic