2017 IBM Research Workshop on Architectures for Cognitive Computing and Datacenters - overview

Cognitive Meter

The 2017 IBM Research Workshop on Architectures for Cognitive Computing and Datacenters will be held on Thursday October 19th, 2017 at the IBM Thomas J. Watson Research Center in Yorktown Heights, New York.

Livestream link: https://bluestream.watson.ibm.com/v4/yorktown/

Presentation slides: IBMBox Folder

This event is a full-day workshop that provides a forum for invited students in the fields of cognitive computing, datacenter, computer architecture, and computer system design to showcase their work and interact with their peers and members of the IBM Research community.

Detailed Agenda

8:30 9:00 Breakfast
9:00 10:00 Keynote talk
Hillery Hunter - Director, Accelerated Cognitive Infrastructure, IBM Fellow
Talk title: Creating Cognitive Capabilities -- How Can Systems Help?
10:00 10:30 Ritchie Zhao - Cornell University, advised by Prof. Zhiru Zhang
Talk title: Accelerating Binarized Neural Networks with Software-Programmable FPGAs
10:30 11:00 Patrick Judd - University of Toronto, advised by Prof. Andreas Moshovos
Talk title: Bit-Pragmatic Deep Neural Network Computing
11:00 11:15 Break
11:15 11:45 Bert Moons - KU Leuven, advised by Prof. Marian Verhelst
Talk title: Algorithmic-, Processor-, and Circuit-techniques for Embedded Deep Learning
11:45 12:15 Hari Cherupalli - University of Minnesota, advised by Prof. John Sartori
Talk title: Application-specific Design and Optimization for Ultra-Low-Power Embedded Systems
12:15 1:00 Lunch
1:00 2:00 Quantum Lab Tour & Break
2:00 2:30 Ozan Tuncer - Boston University, advised by Prof. Ayse Coskun
Talk title: Analytics for Detection of Performance and Configuration Anomalies in Large-scale Systems
2:30 3:00 Apoorve Mohan - Northeastern University, advised by Gene Cooperman
Talk title: Marrying HPC and Cloud for Long-Term Happiness
3:00 3:30 Break
3:30 4:00 Kevin Hsieh - Carnegie Mellon University, advised by Prof. Phillip Gibbons and Prof. Onur Mutlu
Talk title: Gaia: Geo-Distributed Machine Learning Approaching LAN Speeds
4:00 4:30 David G. Marquez - University of Buenos Aires, advised by Adrian Cristal (BSC) and Esteban Mocskos (UBA)
Talk title: Towards Efficient Parallelization Hardware-Software Co-Designed Solutions
4:30 5:30 Panel Session led by Seetharami SeelamEvolving Architectures and Systems for Cognitive Applications
Paul Crumley
Liana Fong
Anthony Hylick
Prashant Nair
Ravi Nair
Jinjun Xiong
5:30   Closing Remarks
6:00   Student Dinner

Instructions for visiting the T. J. Watson Research Center can be found here.

For more information please contact:
Michael B. Healy - mbhealy@us.ibm.com
Zehra Sura - zsura@us.ibm.com
Augusto Vega - ajvega@us.ibm.com

IBM Cognitive THINK