Abstract: Cyber-physical systems (CPS), such as robots and self-driving cars, demand rigorous scheduling to prevent failure, where every millisecond of processing time can be critical. These systems often rely on heterogeneous computing environments, which include CPUs, GPUs, and specialized accelerators, to meet their computational needs efficiently. However, leveraging these diverse processing units to fulfill strict physical constraints remains a significant challenge, as existing scheduling solutions often fall short of addressing the complexities involved in a comprehensive manner. This talk by Dr. Mehmet E. Belviranli delves into the intricate world of creating efficient compute schedules for CPS that not only cater to their diverse computational hardware but also adhere to real-world constraints critical for system safety. We begin by examining the role of neural network (NN) inference in CPS, exploring strategies to balance energy consumption, latency and throughput by distributing the layers of NN across different accelerators. We then introduce a novel, end-to-end framework that integrates physical constraints, heterogeneous computational resources, and latency considerations into a cohesive mixed-integer linear problem, demonstrating through case studies how this approach yields optimal scheduling solutions under varied conditions. Through this exploration, we aim to shed light on the untapped potential of heterogeneous computing in enhancing the reliability and performance of CPS. We will also outline future directions in developing a more robust ecosystem for these complex computing environments, highlighting our contribution to this evolving field. Speaker(s): Dr. Belviranli Virtual: https://events.vtools.ieee.org/m/417666

Join us Thursday, 2 May, for a Special IEEE-USA Washington Update to learn about how the Creating Resources for Every American To Experiment with Artificial Intelligence Act of 2023 (CREATE AI Act) shapes AI development. Our special guest, Katerina (Katie) Antypas from the National Science Foundation, will delve into the establishment of the National Artificial Intelligence Research Resource (NAIRR)—a vital hub for AI researchers and students. Speaker(s): Katerina (Katie) Antypas Virtual: https://events.vtools.ieee.org/m/417483

IEEE ComSoc Norther Virginia chapter and GMU Department of Computer Science invites you to attend the following Distinguished Lecture: Title: Why do small language models underperform? Speaker: Benoît Sagot, Director of Research at INRIA Date: May 2, 2024 Time: 11:00am – 12:00pm In person Location: GMU Fairfax campus, Nguyen Engineering Bldg., Conference Room 4201 Virtual: Microsoft Teams: (https://nam11.safelinks.protection.outlook.com/ap/t-59584e83/?url=https%3A%2F%2Fteams.microsoft.com%2Fl%2Fmeetup-join%2F19%253ameeting_OGUwZGI0OTktNTdhZS00NmNlLWEzZGEtZTJhNGI2Yjg5YmJi%2540thread.v2%2F0%3Fcontext%3D%257b%2522Tid%2522%253a%25229e857255-df57-4c47-a0c0-0546460380cb%2522%252c%2522Oid%2522%253a%2522f9586db0-74ee-4635-a01b-2383b74f8a0c%2522%257d&data=05%7C02%7Ckhassan1%40gmu.edu%7Cc082474ca0814691061508dc69ddedfd%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C638501649141618849%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=MUjIg2xib2wJ22N21llS60nE%2FSTEPpqg%2FSVOdLXJFHA%3D&reserved=0) Meeting ID: 292 789 339 112 Passcode: jM8w7c --------------------------------------------------------------- Dial-in by phone (tel:+15713972084,,218888141#) United States, Arlington (https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdialin.teams.microsoft.com%2F9424c9fe-3b57-41d6-9131-1d3b9b7cf4a9%3Fid%3D218888141&data=05%7C02%7Ckhassan1%40gmu.edu%7Cc082474ca0814691061508dc69ddedfd%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C638501649141626316%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=r6YY83JURqPLJf9J2QcdQ5H6w3QBZTlOs4HAUUBvvSQ%3D&reserved=0) Phone conference ID: 218 888 141# For organizers: (https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fteams.microsoft.com%2FmeetingOptions%2F%3ForganizerId%3Df9586db0-74ee-4635-a01b-2383b74f8a0c%26tenantId%3D9e857255-df57-4c47-a0c0-0546460380cb%26threadId%3D19_meeting_OGUwZGI0OTktNTdhZS00NmNlLWEzZGEtZTJhNGI2Yjg5YmJi%40thread.v2%26messageId%3D0%26language%3Den-US&data=05%7C02%7Ckhassan1%40gmu.edu%7Cc082474ca0814691061508dc69ddedfd%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C638501649141633701%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=KSJGsLYD3jKk7GBpn2XuSPetsCGuvQn9I6sJYXPZfs0%3D&reserved=0) | (https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdialin.teams.microsoft.com%2Fusp%2Fpstnconferencing&data=05%7C02%7Ckhassan1%40gmu.edu%7Cc082474ca0814691061508dc69ddedfd%7C9e857255df574c47a0c00546460380cb%7C0%7C0%7C638501649141641487%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=2ONDG7yjY3RckniAVPkRkliPHOBWdNqIhM2RgAA8O10%3D&reserved=0) Abstract: Language models, and in particular generative and conversational language models, are at the heart of recent advances in natural language processing (NLP). Understanding how these models represent textual content and how they learn these representations still raises multiple research questions. In this talk, I will start from an observation that small models are less efficient than expected. I will show that language models relying on the Transformer architecture tend to produce vector representations that are not isotropically distributed in space. This anisotropy is linked to the way in which these models are learned, which leads to the frequency of the tokens taking a preponderant place in their representation. I will show that this effect has negative consequences on the ability of small models to train satisfactorily (“performance saturation”) but does not seem to affect larger models. I will then describe a new approach for training language models intended to avoid the undesirable effects of this prevalence of frequency information. The resulting “headless” models display a number of advantages over standard models, including on downstream performance. Bio: Benoît Sagot is a computer scientist specialized in natural language processing (NLP). He is a Senior Researcher (Directeur de Recherches) at INRIA, where is heads the INRIA research project ALMAnaCH in Paris, France. He also holds a chair in the PRAIRIE institute dedicated to artificial intelligence, and currently holds the annual chair for computer science in the Collège de France. His research focuses on language modelling, machine translation, language resource development and computational linguistics, with a focus on French in all its form and on less-resourced languages. ________________________________________________________________________________ Co-sponsored by: GMU Department of Computer Science Bldg: Nguyen Engineering Bldg., Conference Room 4201, 4400 University Drive , Fairfax, Virginia, United States, 22030, Virtual: https://events.vtools.ieee.org/m/419402

Normally scheduled ExCom meeting for the 1st Thursday of each month. Agenda: TBD Bldg: MAI, (Checkpoint Surgical) 6050 Oak Tree Blvd, Independence, Ohio, United States, Virtual: https://events.vtools.ieee.org/m/417704

The Columbus, OH Section Joint Chapter (SSC37/CAS04) is excited to announce its return in 2024 with a dynamic lineup of workshops! Thanks to the generous sponsorship from the CAS society, the chapter is proud to present a new series titled "Foundations of Mixed-Signal IC Design: A Practical Approach to Lab-to-Fab." This series will offer a unique blend of lecture-style talks delivered by subject matter experts in the field of integrated circuit design, along with hands-on technical sessions. These sessions will guide both students and professionals through the digital design flow, preparing them for submission on one of the (https://tinytapeout.com/) shuttle runs! The society is delighted to sponsor 15-20 enthusiastic students and professionals who are eager to participate in the (https://tinytapeout.com/) shuttles at the end of the year. The (https://tinytapeout.com/) workshop series will kick off with the first of three sessions aimed at equipping participants with essential digital design fundamentals and familiarizing them with the open-source digital tool flow. The inaugural workshop will present foundational material derived from workshop slides provided by Tiny Tapeout (https://tinytapeout.com/teaching/). Topics covered will include semiconductor fabrication, historical perspectives on process scaling, and the fundamentals of digital design flow using open-source tools. Following the informational session, participants will receive resources to set up the development tools on their personal computers, enabling them to embark on their own experimentation and design endeavors. Subsequent workshops will delve into more advanced applications of the digital tool flow, as well as interactive troubleshooting sessions addressing participants' design challenges. Based on feedback received, additional in-person or remote support sessions may be introduced. These workshops will complement lectures delivered by subject matter experts on IC design and testing. By the conclusion of the three sessions, our objective as society leaders is to furnish participants with a solid understanding of integrated circuit design, fabrication, and testing, with a particular emphasis on leveraging the open-source digital flow offered by the Tiny Tapeout shuttle program This initiative presents a remarkable opportunity for hands-on learning in VLSI design within a supportive environment conducive to skill development. Google Meet link: https://meet.google.com/yvu-womd-jni 1275 Kinnear Rd, Columbus, Ohio, United States, 43212, Virtual: https://events.vtools.ieee.org/m/416616