Artificial intelligence

MIT researchers have developed a new robot that can help locate hidden items using AI and wireless technologies, reports Benoit Morenne for The Wall Street Journal. “The latest version of the robot has a 96% success rate at finding and picking up objects in a lab setting, including clothes and household items,” writes Morenne. “In the future, this home helper could also retrieve a specific wrench or screwdriver from a toolbox and assist a human in assembling a piece of furniture.”

Koch Institute fellow Dr. Rameen Shakur and his colleagues have developed a new computer tool that could allow doctors to personalize treatments for patients with inherited heart disease. “In areas such as cardiology and oncology, where large amounts of clinical and genetic data need to be analyzed, adopting a computer-based approach…can make diagnosis, outcome prediction and treatment more effective and efficient,” writes Helen Albert for Clinical OMICs.

MIT researchers have developed new programmable fibers that could help transform clothing into wearable computers, reports Kyle Mizokami for Popular Mechanics. “The polymer fibers contain hundreds of tiny silicon microchips that, once electrified, can sustain a digital connection across tens of meters,” Mizokami writes.

Forbes contributor Eric Tegler spotlights how MIT researchers are developing a fiber with digital capabilities. “Individuals wearing garments with digital fibers could be alerted to vital information about their physiology and environmental exposures, and share health/injury and location data with support forces,” Tegler explains.

Kate Darling, a research specialist at the Media Lab, speaks with Ira Flatow on Science Friday about “the historical lessons of our relationships with animals, and how they could inform our legal, ethical, and even emotional choices about robots and AI.”

Mashable spotlights Strolling Cities, a video project from the MIT-IBM Watson AI Lab, which uses AI to allow users to imagine what different words would like as a location. “Unlike other image-generating AI systems, Strolling Cities creates fictional cities every time,” Mashable notes.

UPI reporter Brooks Hays writes that researchers from MIT and other institutions have developed a programmable digital fiber that can capture, store and analyze data. The technology could “be paired with machine learning algorithms and used to make smart fabrics to record health data and aid medical diagnosis,” writes Hays.

An electric, autonomous boat developed by MIT researchers is being tested in the canals of Amsterdam as part of an effort to ease traffic, reports Aleksandar Furtula and Mike Corder for the AP. The Roboat project is aimed at developing “new ways of navigating the world’s waterways without a human hand at the wheel,” write Furtula and Corder. “The vessels are modular so they can be easily adapted for different purposes, carrying cargo or workers.”

A recent review by MIT researchers finds that “only about 23% of machine learning studies in health care used multiple datasets to establish their results, compared to 80% in the adjacent field of computer vision, and 58% in natural language processing,” writes Casey Ross for STAT. “If the performance results are not reproduced in clinical care to the standard that was used during [a study], then we risk approving algorithms that we can’t trust,” says graduate student Matthew McDermott. “They may actually end up worsening patient care.”

Fast Company reporter Mark Wilson spotlights Strolling Cities, a new AI video project developed by researchers from the MIT-IBM Watson AI Lab, which recreates the streets of Italy based on millions of photos and words. “I decided that the beauty and sentiment, the social, historical, and psychological contents of my memories of Italy could become an artistic project, probably a form of emotional consolation,” says Mauro Martino of the MIT-IBM Watson AI Lab. “Something beautiful always comes out of nostalgia.”

Fortune reporter Jonathan Vanian writes that researchers from the MIT-IBM Watson AI Lab and IBM Research have published a paper and data set “intended to help researchers create deep learning systems that can create software code.” Vanian notes that “research like this could pave the way for more efficient methods for companies to test software, analyze code, and potentially build apps.”

Times Higher Education reporter Simon Baker writes that Media Lab researchers have developed a new machine learning model that can predict research studies that will have the highest impact. The tool has the potential to “aid funders and research evaluators in making better decisions and avoiding the kind of biases and gaming that occurred with simpler metric assessments.”

Wired reporter Will Knight spotlights how MIT researchers have showed that “an AI program trained to verify that code will run safely can be deceived by making a few careful changes, like substituting certain variables, to create a harmful program.”

In an article for New Scientist, Vijaysree Venkatraman reviews a new book by Kate Darling of the Media Lab, which explores whether we should think of robots as more like animals than humans. “Unlike animals, robots are designed, peddled and controlled by people, Darling reminds us. Her timely book urges us to focus on the legal, ethical and social issues regarding consumer robotics to make sure the robotic future works well for all of us,” writes Venkatraman.

 

Graduate student Joy Buolamwini joins Kara Swisher on The New York Times' “Sway” podcast to discuss her crusade against bias in facial recognition technologies. “If you have a face, you have a place in this conversation,” says Buolamwini.