Researchers at the University of Pennsylvania propose a new computing architecture ideal for artificial intelligence Penn In the News MarkTechPost Researchers at the University of Pennsylvania propose a new computing architecture ideal for artificial intelligence A team of researchers at the School of Engineering and Applied Science and colleagues have created a new computing architecture based on compute-in-memory, which is ideal for AI. What’s in the semiconductor bill? Q&A What’s in the semiconductor bill? In a Q&A, Morris Cohen of the Wharton School explains the content of the CHIPS and Science Act, signed into law on Aug. 9. PIK Professor Kevin Johnson: Informatics evangelist PIK Professor Kevin Johnson: Informatics evangelist The Penn Integrates Knowledge Professor with appointments in Penn Engineering and the Perelman School of Medicine on forging his own path in the fields of health care and computer science. Penn Abroad: Rising senior Ricardo Del Rio in Switzerland Rising senior Ricardo Del Rio -- an electrical engineering major in the School of Engineering and Applied Science from Guadalajara, Mexico -- studied abroad at ETH Zurich in Switzerland during the spring semester. Penn Abroad: Rising senior Ricardo Del Rio in Switzerland Rising senior Ricardo Del Rio, an electrical engineering major in the School of Engineering and Applied Science, studied at ETH Zurich during the spring semester. A chip that can classify nearly two billion images per second Using a deep neural network of optical waveguides, the researchers’ chip—smaller than a square centimeter—can detect and classify an image in less than a nanosecond, all without the need for a separate processor or memory unit. (Image: Ella Maru Studio/Penn Engineering Today) A chip that can classify nearly two billion images per second Using a deep neural network of optical waveguides, a new chip developed by Penn engineers—smaller than a square centimeter—can detect and classify an image in less than a nanosecond, all without the need for a separate processor or memory unit. Discovering new ways to control light The different colors in this sample of iron phosphorous trisulfide (FePS3) correspond to regions with varying thicknesses, which form different “cavity” modes at different wavelengths. (Image: Penn Engineering Today) Discovering new ways to control light Researchers found a magnetic property in a class of materials that enables light manipulation on the nanoscale, with implications for applications such as information storage and energy harvesting. How to design a sail that won’t tear or melt on an interstellar voyage Artist rendering of the Starshot Lightsail spacecraft during acceleration by a ground-based laser array. Previous conceptions of lightsails have imagined them being passively pushed by light from the sun, but Starshot’s laser-based approach requires rethinking the sail’s shape and composition so it won’t melt or tear during acceleration. (Image: Masumi Shibata, courtesy of Breakthrough Initiatives) How to design a sail that won’t tear or melt on an interstellar voyage The Breakthrough Starshot Initiative’s laser-based approach requires rethinking a sail’s shape and composition so it won’t melt or tear during acceleration and pushed by wind, not light. Refining data into knowledge, turning knowledge into action Homepage image: No one type of medical imaging can capture every relevant piece of information about a patient at once. Digital twins, or multiscale, physics-based simulations of biological systems, would allow clinicians to accurately infer more vital statistics from fewer data points. Refining data into knowledge, turning knowledge into action Penn Engineering researchers are using data science to answer fundamental questions that challenge the globe—from genetics to materials design. New atomically-thin material could improve efficiency of light-based tech So-called “two-dimensional” materials have unique electrical and photonic properties, but their ultrathin form factors present practical challenges when incorporated into devices. Penn Engineering researchers have now demonstrated a method for making large-area “superlattices”—layered structures containing 2D lattices of sulfur and tungsten—that can achieve light-matter coupling. (Image: Penn Engineering Today) New atomically-thin material could improve efficiency of light-based tech A new photodetector design from Penn Engineering is not only extremely thin, making it lightweight and cost effective, it can also emit light, not just detect it. Putting community first, in a new social justice initiative Q&A Putting community first, in a new social justice initiative The Community Collaboratory for Co-Creation, led by Penn Nursing and Penn Engineering, will focus on research, education, and community engagement and outreach.
What’s in the semiconductor bill? Q&A What’s in the semiconductor bill? In a Q&A, Morris Cohen of the Wharton School explains the content of the CHIPS and Science Act, signed into law on Aug. 9.
PIK Professor Kevin Johnson: Informatics evangelist PIK Professor Kevin Johnson: Informatics evangelist The Penn Integrates Knowledge Professor with appointments in Penn Engineering and the Perelman School of Medicine on forging his own path in the fields of health care and computer science.
Penn Abroad: Rising senior Ricardo Del Rio in Switzerland Rising senior Ricardo Del Rio -- an electrical engineering major in the School of Engineering and Applied Science from Guadalajara, Mexico -- studied abroad at ETH Zurich in Switzerland during the spring semester. Penn Abroad: Rising senior Ricardo Del Rio in Switzerland Rising senior Ricardo Del Rio, an electrical engineering major in the School of Engineering and Applied Science, studied at ETH Zurich during the spring semester.
A chip that can classify nearly two billion images per second Using a deep neural network of optical waveguides, the researchers’ chip—smaller than a square centimeter—can detect and classify an image in less than a nanosecond, all without the need for a separate processor or memory unit. (Image: Ella Maru Studio/Penn Engineering Today) A chip that can classify nearly two billion images per second Using a deep neural network of optical waveguides, a new chip developed by Penn engineers—smaller than a square centimeter—can detect and classify an image in less than a nanosecond, all without the need for a separate processor or memory unit.
Discovering new ways to control light The different colors in this sample of iron phosphorous trisulfide (FePS3) correspond to regions with varying thicknesses, which form different “cavity” modes at different wavelengths. (Image: Penn Engineering Today) Discovering new ways to control light Researchers found a magnetic property in a class of materials that enables light manipulation on the nanoscale, with implications for applications such as information storage and energy harvesting.
How to design a sail that won’t tear or melt on an interstellar voyage Artist rendering of the Starshot Lightsail spacecraft during acceleration by a ground-based laser array. Previous conceptions of lightsails have imagined them being passively pushed by light from the sun, but Starshot’s laser-based approach requires rethinking the sail’s shape and composition so it won’t melt or tear during acceleration. (Image: Masumi Shibata, courtesy of Breakthrough Initiatives) How to design a sail that won’t tear or melt on an interstellar voyage The Breakthrough Starshot Initiative’s laser-based approach requires rethinking a sail’s shape and composition so it won’t melt or tear during acceleration and pushed by wind, not light.
Refining data into knowledge, turning knowledge into action Homepage image: No one type of medical imaging can capture every relevant piece of information about a patient at once. Digital twins, or multiscale, physics-based simulations of biological systems, would allow clinicians to accurately infer more vital statistics from fewer data points. Refining data into knowledge, turning knowledge into action Penn Engineering researchers are using data science to answer fundamental questions that challenge the globe—from genetics to materials design.
New atomically-thin material could improve efficiency of light-based tech So-called “two-dimensional” materials have unique electrical and photonic properties, but their ultrathin form factors present practical challenges when incorporated into devices. Penn Engineering researchers have now demonstrated a method for making large-area “superlattices”—layered structures containing 2D lattices of sulfur and tungsten—that can achieve light-matter coupling. (Image: Penn Engineering Today) New atomically-thin material could improve efficiency of light-based tech A new photodetector design from Penn Engineering is not only extremely thin, making it lightweight and cost effective, it can also emit light, not just detect it.
Putting community first, in a new social justice initiative Q&A Putting community first, in a new social justice initiative The Community Collaboratory for Co-Creation, led by Penn Nursing and Penn Engineering, will focus on research, education, and community engagement and outreach.