News

Linh Thi Xuan Phan has been appointed the co-director of the Data Science (DATS) Master's program. 

Linh is widely known as a dedicated teacher and someone highly engaged with master's students in education and research.  Many DATS students are taking her popular and highly-rated CIS 505, which covers topics related to big data infrastructures and cloud computing. She will add new dimensions to the program, given her expertise in big data infrastructures, cloud computing, and enhance DATS connections to embedded (health) device data analytics.

Congratulations, Linh!

We are pleased to welcome Justin Gottschlich, a respected and well-known leader in the industry, to PRECISE.  He has been appointed Industrial Board Chair and Executive Director of AI R&D.

Meng Xu, Linh Thi Xuan Phan, and Hyon-Young Choi (University of Pennsylvania); Yuhan Lin (Northeastern University); Haoran Li and Chenyang Lu (Washington University in St. Louis); and Insup Lee (University of Pennsylvania) are the recipients of the Best Paper Award at the 25th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), part of the Cyber-Physical Systems Week (CPSWeek), that took place in Montreal, Canada on 15-18 April 2019.  Their paper titled “Holistic Resource Allocation for Multicore Real-Time Systems” presents a holistic cache and memory bandwidth resource allocation strategy for multicore real-time systems.  Their strategy exploits the relationship between the allocation of cache and memory bandwidth resources and a task's WCET to map tasks onto cores and to compute the resource allocation for each core, to fully utilize resources while ensuring timing guarantees.  Extensive evaluations using real-world benchmarks show that their strategy offers near optimal schedulability performance while being highly efficient, and that it substantially outperforms state-of-the-art solutions.

Linh, and her former/current doctoral students (Saeed Abedi, Neeraj Gandhi, Henri Maxime Demoulin, Yang Li, and Yang Wu), also won RTAS Best Student Paper Award.  Their paper titled “RTNF: Predictable Latency for Network Function Virtualization” presents a scalable framework for the online resource allocation and scheduling of NFV applications that provides predictable end-to-end latency guarantees.  RTNF is based on a novel time-aware abstraction algorithm that transforms complex NFV graphs and their performance requirements into sets of scheduling interfaces; these can then be used by the resource manager and the scheduler on each node to efficiently allocate resources and to schedule NFV requests at runtime.  Their evaluation, based on simulations and an experimental prototype, shows that RTNF can schedule DAG-based NFV applications with solid timing guarantees while incurring only a small overhead, and that it substantially outperforms state-of-the-art  techniques.

Congratulations to all!

Verisig is a tool for verifying properties of neural networks in autonomous systems.  The novelty of Verisig lies in its encoding of a deep neural networks as hybrid systems such that it can be easily composed with hybrid systems models of vehicle dynamics and verified using state-of-the-art solvers (e.g., Flow*).  Consequently, Verisig has been used to verify safety properties of learning-enabled closed-loop controllers containing neural networks with 10s of layers and 100s of neurons per layer.  Verisig 0.9 represents the first public release of the tool, being actively developed as part of the DARPA Assured Autonomy program.

Cyberjutsu Girls (@CyberjutsuGirls) tweeted at 1:19 PM on Sat, Mar 09, 2019:
Today we're learning how cyber & medicine overlap in a new biomedical workshop: "Music from the Heart!" Our ladies are using @adafruit Huzzah to monitor heart rates & compete in a "rock band challenge" to see who can control them! Thanks to @IoMTprof for instructing! #GirlsInSTEM https://t.co/Rp20gT2adP

The Deans' Distinguished Visiting Professorship is award to Insup Lee, Ph.D. by Perelman School of Medicine at the University of Pennsylvania on 17 Jan 2019. Dr. Lee presented to the audience a talk entitled "Internet of Medical Things (IoMT)" that day.

Micelio is Shreyas Shibulal’s new early-stage fund which will also be building a design discovery studio apart from investing in startups in the electric vehicle space.  

Congratulations to Grayson Honan!  He is awarded the ESE BEST TEACHING ASSISTANT AWARD for the 2017-2018 Academic Year.  Teaching is a task that says a lot about who we are. Throughout his two-year academic career, Grayson has always went out of his way to help those who need it.  Thank you, Grayson, for working diligently to ensure the next generation of engineers are posed to make a difference.

Nimit Singhania won the Radhia Cousot Young Researcher Best Paper Award at the 25th Static Analysis Symposium (SAS 2018) that took place in Freiburg im Breisgau, Germany on 29-31 August 2018. In his paper titled "Block-Size Independence for GPU Programs" (joint work with his PhD advisors Rajeev Alur and Joseph Devietti), he proposes a new property called "block-size independence", and an accompanying compiler analysis, which guarantees that adjusting the block-size of a GPU program does not change what it computes; but tuning the block-size can result in significant performance speedups, especially as code is moved to different kinds of GPUs.

Dagaen Golomb, Deepak Gangadharan, Sanjian Chen, Oleg Sokolsky and Insup Lee are the recipients of the Best Paper Award at the 21st IEEE International Symposium on Real-Time Computing (ISORC 2018) that took place in Singapore on 29-31 May 2018. In their paper titled “Data Freshness Over Engineering: Formulation and Results”, they propose a method to determine the periods of tasks in chains of arbitrary length while satisfying end-to-end freshness constraints with only few assumptions regarding the scheduling algorithm used.