ERIK MCLEAN / UNSPLASH

Friday, 27. May 2022 5:00 pm  Photonic computing beyond MooreÂ’s Law

Prof. Dr. Wolfram Pernice, Kirchoff-Institut für Physik, Universität Heidelberg Ever noticed that annoying lag that sometimes happens during the internet streaming from, say, your favorite football game? Called latency, this brief delay between a camera capturing an event and the event being shown to viewers is surely annoying during the decisive goal at a World Cup final. But it could be deadly for a passenger of a self-driving car that detects an object on the road ahead and sends images to the cloud for processing. A way to dramatically reduce latency in artificial intelligence (AI) systems lies in using light for computation instead of electronic circuits. Combining photonic processing with what’s known as the non-von Neumann, in-memory computing paradigm enables to perform computations with unprecedented, ultra-low latency and compute density. Photonic tensor cores run computations at a processing speed higher than ever before and perform key computational primitives associated with AI models such as deep neural networks for computer vision, with remarkable areal and energy efficiency. While scientists first started tinkering with photonic processors back in the 1950s, in-memory computing (IMC) is an emerging non-von Neumann compute paradigm where memory devices, organized in a computational memory unit, are used for both processing and memory. By removing the need to shuffle data around between memory and processing units, IMC even with conventional electronic memory devices could bring significant latency gains. However, the combination of photonics with IMC could further reduce the latency issue – so efficiently that photonic in-memory computing might soon play a key role in latency-critical AI applications. Together with in-memory computing, photonic processing overcomes the seemingly insurmountable barrier to the bandwidth of conventional AI computing systems based on electronic processors.

Astronomy colloquium

Tuesday, 24. May 2022 4:00 pm  Stellar feedback: from stars to galaxies

Dr Anna McLeod, Centre for Extragalactic Astronomy, Durham University, UK Feedback from massive stars plays a central role in shaping the evolution of galaxies. Conversely, different galactic environments play a central role in regulating the impact of massive stars. Yet, despite a solid qualitative understanding of feedback, our quantitative knowledge about the interdependence of feedback and environment remains poor. Until recently, only a small number of star-forming regions had adequate observational information on both gas and stars needed for detailed stellar feedback studies. Over the past decade, integral field units (IFUs) have revolutionized our approach to resolved stellar feedback studies in nearby galaxies. In this talk I will present recent results of IFU nearby galaxy surveys, showcasing how these can be used to simultaneously characterize the feedback-driven interstellar medium and individual feedback-driving stars up to Mpc distances, and I will discuss how this enables the first empirical quantification of the interdependence between stellar feedback and the environments massive stars form in. Lastly, if there is time, I will also be talking about how IFU data can lead to truly serendipitous discoveries. Dr. McLeod will be based at the Astronomisches Rechen-Institut for her visit to Heidelberg and will be available for meetings by arrangement with her host, Melanie Chevance (chevance@uni-heidelberg.de). Those unable to attend the colloquium in person are invited to participate online through Zoom (Meeting ID: 942 0262 2849, passcode 792771) using the link: https://zoom.us/j/94202622849pwd=dGlPQXBiUytzY1M2UE5oUDRhbzNOZz09

Center for Quantum Dynamics Colloquium

Wednesday, 8. June 2022 5:00 pm  tba

Prof. Dr. Ludwig Mathey, Institut für Laserphysik, Universität Hamburg