Meeting updates joint research projects between the two institutions and will feature 'genius grant' recipient Prof. Regina Barzilay, who uses AI to detect cancer, as keynote speaker.
Researchers from the Massachussetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory (MIT-CSAIL) are visiting Doha this week as part of a collaboration with QCRI.
The meeting will focus on updating recent projects between the two institutions, who have been conducting computer science research together for almost six years.
“Our collaboration with QCRI has been very rewarding for our laboratory in a rich myriad of ways,” says CSAIL Director Daniela Rus (pictured). “We have done a wide range of projects together that have already had a major impact in the field of computer science.”
A keynote speaker at the event will be CSAIL’s Regina Barzilay, a recent MacArthur “genius” who develops artificial intelligence systems aimed at better detecting cancer. Prof. Barzilay’s talk, on Tuesday, March 27 at 4 p.m at the Education City Student Center, is open to the public.
Prof. Barzilay will discuss how machine learning models have been used to read imaging data, analyze patient records, and monitor how diseases progress over time.
The QCRI-CSAIL Computer Science Research Program has focused on a diverse range of topics, from database management and speech processing to shared computing and data analytics, including:
• Pic2Recipe, an AI that can look at a photo of a dish and be able to suggest similar recipes
• A project focused on developing accurate map-making techniques via crowd-sourcing
• Data Civilizer, a system for easier data-filtering that finds and linking information scattered across digital files
• Interactive Dynamic Video, a new technique that lets you reach in and “touch” objects in videos using a normal camera
• Better speech and language processing technologies, particularly for addressing low-resource issues related to the Arabic language
• A paper in which researchers showed that they had been able to expose a vulnerability in Tor, the anonymity-minded network of roughly 2.5 million users
• 2D-to-3D, a system that automatically converts 2-D video of sporting events into 3-D content that can be played back over any 3-D device
The annual project review is open to the public on Tuesday, March 27. To register for this event, please click here
To register for Prof. Barzilay’s talk, please click here.