Algorithms

MIT researchers have developed a “risk-aware” model that could improve the performance of cloud-computing networks across the U.S.

Using Wall Street secrets to reduce the cost of cloud infrastructure

“Risk-aware” traffic engineering could help service providers such as Microsoft, Amazon, and Google better utilize network infrastructure.

August 18, 2019

A new MIT-developed model automates a critical step in using AI for medical decision making, where experts usually identify important features in massive patient datasets by hand. The model was able to automatically identify voicing patterns of people with vocal cord nodules (shown here) and, in turn, use those features to predict which people do and don’t have the disorder.

Automating artificial intelligence for medical decision-making

Model replaces the laborious process of annotating massive patient datasets by hand.

August 5, 2019

A model developed at MIT predicts the cognitive decline of patients at risk for Alzheimer’s disease by forecasting their cognition test scores up to two years in the future, which could help zero in on the right patients to select for clinical trials.

Model predicts cognitive decline due to Alzheimer’s, up to two years out

Researchers hope the system can zero in on the right patients to enroll in clinical trials, to speed discovery of drug treatments.

August 1, 2019

MIT graduate students Logan Engstrom (left) and Andrew Ilyas have published five papers together exploring the vulnerabilities of computer vision systems, and they have half a dozen more in the pipeline.

Why did my classifier just mistake a turtle for a rifle?

Two longtime friends explore how computer vision systems go awry.

July 31, 2019

A course called Machine Learning for Healthcare introduces students to the technical and ethical challenges of using automated tools to diagnose and treat patients.

Want to know what software-driven health care looks like? This class offers some clues.

A course that combines machine learning and health care explores the promise of applying artificial intelligence to medicine.

July 24, 2019

“I hope to save a lot of people a lot of time doing boring, repetitive work by automating programming work as well as decreasing the cost of building secure, reliable systems,” says Associate Professor Adam Chlipala.

Making it easier to program and protect the web

Professor Adam Chlipala builds tools to help programmers more quickly generate optimized, secure code.

July 20, 2019

A new MIT-invented system automatically designs and 3-D prints complex robotic actuators optimized according to an enormous number of specifications, such as appearance and flexibility. To demonstrate the system, the researchers fabricated floating water lilies with petals equipped with arrays of actuators and hinges that fold up in response to magnetic fields run through conductive fluids.

Automated system generates robotic parts for novel tasks

When designing actuators involves too many variables for humans to test by hand, this system can step in.

July 12, 2019

Omer Tanovic says that his engineering background has taught him never to lose sight of the intended applications of his work, or the practical parameters for implementation.

Making wireless communication more energy efficient

Along with studying theory, "it's also important to me that the work we are doing will help to solve real-world problems,” says LIDS student Omer Tanovic.

July 3, 2019

The GANpaint system developed at MIT can easily add features to an existing image. At left, the original photo of a kitchen; at right, the same kitchen with the addition of a window. Co-author Jun-Yan Zhu believes better understanding of GANs will help researchers be able to better stamp out fakery: “This understanding may potentially help us detect fake images more easily.”

Teaching artificial intelligence to create visuals with more common sense

An MIT/IBM system could help artists and designers make quick tweaks to visuals while also helping researchers identify “fake” images.

July 1, 2019

For years, researchers from MIT and Brown University have been developing an interactive system that lets users drag-and-drop and manipulate data on any touchscreen, including smartphones and interactive whiteboards. Now, they’ve included a tool that instantly and automatically generates machine-learning models to run prediction tasks on that data.

Drag-and-drop data analytics

System lets nonspecialists use machine-learning models to make predictions for medical research, sales, and more.

June 27, 2019

With the IdPrism platform, algorithms encode genetic markers as bits to allow for fast DNA comparisons in forensics. “When Hollywood finds out about this, they’re going to change their scripts,” says Darrell Ricke of the Bioengineering Systems and Technologies Group. “The capabilities are so different from what’s out there.”

Record-breaking DNA comparisons drive fast forensics

Developed at MIT Lincoln Laboratory, IdPrism and its award-winning algorithms provide rapid analysis for complex forensic DNA samples.

June 26, 2019

Users feed Gen relatively short code defining a target task, and the system automatically generates the results.

New AI programming language goes beyond deep learning

General-purpose language works for computer vision, robotics, statistics, and more.

June 26, 2019

The MIT Machine Intelligence Community board and executive committee

Want to learn how to train an artificial intelligence model? Ask a friend.

MIT Machine Intelligence Community introduces students to nuts and bolts of machine learning.

June 25, 2019

Robots currently attempt to identify objects in a point cloud by comparing a template object — a 3-D dot representation of an object, such as a rabbit — with a point cloud representation of the real world that may contain that object.

Spotting objects amid clutter

New approach quickly finds hidden objects in dense point clouds, for use in driverless cars or work spaces with robotic assistants.

June 19, 2019

MIT researchers have developed a system that gleans far more labeled training data from unlabeled data, which could help machine-learning models better detect structural patterns in brain scans associated with neurological diseases. The system learns structural and appearance variations in unlabeled scans, and uses that information to shape and mold one labeled scan into thousands of new, distinct...

From one brain scan, more information for medical artificial intelligence

System helps machine-learning models glean training information for diagnosing and treating brain conditions.