3/14
Computer Science
Training medical AI with knowledge, not shortcuts
Training medical AI with knowledge, not shortcuts
By emulating the training pathway of human physicians, KnoBo breaks new ground in medical image recognition with AI.
The mechanics of collaboration
The mechanics of collaboration
Penn Ph.D. student Xinlan Emily Hu leads a group of budding engineers and social scientists who study communication across teams. The group has developed a new toolkit aimed at helping researchers analyze and measure teamwork.
A mission to supercharge scientific research
A mission to supercharge scientific research
Jacob Gardner, an assistant professor in computer and information science, wants to leverage AI to accelerate scientific research across disciplines.
Recognizing a pioneer: Penn Engineering’s Grace Hopper
Recognizing a pioneer: Penn Engineering’s Grace Hopper
Hopper was honored for developing the A-0 compiler, an early innovation in computer programming.
The key to fixing AI bias and copyright infringement
The key to fixing AI bias and copyright infringement
Penn Engineering’s Michael Kearns, National Center Professor of Management & Technology, questions whether model disgorgement can potentially solve a number of problems related to AI.
Ensuring the safety and security of AI-controlled systems
Ensuring the safety and security of AI-controlled systems
The doctoral candidate in computer and information science has developed new approaches to address challenges in security and safety for modern cyber-physical systems.
Combining the skills of engineering and design
Combining the skills of engineering and design
Janice Kim, a fourth-year student in the College of Arts & Sciences and the School of Engineering and Applied Science, will graduate as the first Penn student to have a dual degree in computer science and design.
Depression in Black people goes unnoticed by AI models analyzing language in social media posts
Depression in Black people goes unnoticed by AI models analyzing language in social media posts
Penn analysis found that models developed to detect depression using language in Facebook posts did not work when applied to Black people.
How common is common sense?
How common is common sense?
Researchers from Penn develop a framework for quantifying common sense, findings address a critical gap in how knowledge is understood.
Penn alum named 2024 Churchill Scholar
Penn alum named 2024 Churchill Scholar
Xander Uyttendaele, a 2023 graduate, is among 16 selected nationwide to receive the scholarship.
In the News
CNN
What officials are doing—and suggesting—to learn more about the possible drone sightings
Dean Vijay Kumar of the School of Engineering and Applied Science says that the main difficulties in detecting and tracking drones stem from their small size, agility, and potential for autonomous operation.
FULL STORY →
CNN
Why you shouldn’t try to shoot down a suspected drone
Dean Vijay Kumar of the School of Engineering and Applied Science says that shooting at potential drones could lead to missed shots, stray bullets, or accidents that result in hazardous outcomes far beyond the intended action.
FULL STORY →
Newsweek
What New Jersey’s mystery drones could actually be, according to experts
Dean Vijay Kumar of the School of Engineering and Applied believes that mystery drones in New Jersey are part a U.S. government operation rather than being sent by a foreign state actor.
FULL STORY →
Scientific American
Grumpy voters want better stories. Not statistics
In a Q&A, PIK Professor Duncan Watts says that U.S. voters ignored Democratic policy in favor of Republican storytelling.
FULL STORY →
Interesting Engineering
Superhuman vision lets robots see through walls, smoke with new LiDAR-like eyes
Mingmin Zhao of the School of Engineering and Applied Science and colleagues are using radio signals to allow robots to “see” beyond traditional sensor limits.
FULL STORY →
The Conversation
Many wealthy members of Congress are descendants of rich slaveholders — new study demonstrates the enduring legacy of slavery
A co-authored study by Ph.D. student Neil Sehgal of the School of Engineering and Applied Science found that legislators who are descendants of slaveholders are significantly wealthier than members of Congress without slaveholder ancestry.
FULL STORY →