RSP 2020 Program

Rapid System Prototyping is be part of ESWEEK'20 and is held this year as a virtual event through Whova platform, as the other associated ESWEEK events.  

Pre-recorded video presentations and PDF versions of each paper will be available through Whova before the workshop. Attendees will be able to post questions to authors or upvote existing questions in the Q&A forums of each paper provided within Whova.

In addition, live technical sessions are proposed on Thursday (Sept 24). These sessions will be short due to the constraint of different time zones. The authors will present an overview of their papers in 3-minute lightning talks at the beginning of the live session. The remainder of the session is dedicated to live Q&A between authors and attendees, including questions previously posted in the Q&A forums as moderated by the Session Chairs. 

Note that attendees have access to all associated ESWEEK events!

The current version of RSP'2020 live technical sessions is presented below.
 

 

September 24th (Thursday)

 

Indicated time is in Central European Summer Time (CEST) zone

14:15 - 14:30  Welcome - Opening Remarks

14:30 - 15:00  Keynote

  • Prof. Sungjoo Yoo, Seoul National University, Quantizing neural networks for ultra-low-precision computation


15:00 - 15:30  Live technical session A - Session Chair: Sungjoo Yoo, Seoul National University

  • Erwan Moreac, El Mehdi Abdali, Francois Berry, Dominique Heller and Jean-Philippe Diguet.
    Hardware-in-the-loop simulation with dynamic partial FPGA reconfiguration applied to computer vision in ROS-based UAV

  • Daniel Gis, Nils Büscher and Christian Haubelt.
    Advanced Debugging Architecture for Smart Inertial Sensors using Sensor-in-the-Loop

  • Erwan Lenormand, Thierry Goubier, Loïc Cudennec and Henri-Pierre Charles.
    A combined fast/cycle accurate simulation tool for reconfigurable accelerator evaluation: application to distributed data management

15:30 - 15:40    Break

15:40 - 16:10    Live technical session B - Session Chair: Kenneth Kent, University of New Brunswick

  • Chen Wu, Virginie Fresse, Benoit Suffran and Hubert Konik.
    Mathematic models based on multiple-criteria decision analysis for tuning industrial CNN in an FPGA computing cluster

  • José Domingues, Fábio Coutinho, Samuel Pereira, Pedro Marques, Hugerles Silva and Arnaldo Oliveira.
    Fast MPSoC Prototyping of a Reconfigurable DU Downlink Transmission Chain for 5G New Radio

  • Jeremy Nadal and Amer Baghdadi.
    FPGA based design and prototyping of efficient 5G QC-LDPC channel decoding

     

16:10 - 16:20    Break

16:20 - 16:50    Live technical session C - Session Chair: Amer Baghdadi, IMT Atlantique/Lab-STICC

  • Jean Bruant, Pierre-Henri Horrein, Olivier Muller, Tristan Groléat and Frédéric Pétrot.
    (System)Verilog to Chisel Translation for Faster Hardware Design

  • Xuzhi Zhang, Narendra Prabhu and Russell Tessier.
    NestedNet: A Container-based Prototyping Tool for Hierarchical Software Defined Networks

  • Seyed Alireza Damghani, Jean-Philippe Legault and Kenneth B. Kent.
    Desired Footprint by Technology Mapping Modification using a Genetic Algorithm in Odin II

     

16:50 - 17:00   Break - End of Workshop


Keynote

Sungjoo Yoo (Seoul National University), "Quantizing neural networks for ultra-low-precision computation"

Bit-width needs to be minimized for efficient neural network design in terms of chip area, code size, and, most importantly, energy efficiency. In this talk, we first review state-of-the-art quantization methods in the industry and academia and introduce our ideas of outlier quantization, precision highway and quantization error fluctuation-aware training, which finally offers 4-bit linear weight/activation quantization of MobileNet v3.

 
Sungjoo Yoo received Ph.D. from Seoul National University in 2000. From 2000 to 2004, he was researcher at system level synthesis (SLS) group, TIMA laboratory, Grenoble France. From 2004 to 2008, he led, as principal engineer, system-level design team at System LSI, Samsung Electronics. From 2008 to 2015, he was associate professor at POSTECH. In 2015, he joined Seoul National University and is now full professor. His current research interests are software/hardware co-design of deep neural networks and machine learning-based optimization of computer architecture