[hpc-announce] CfP: Special Issue "Accelerating (Deep) Neural Networks and Approximate Computing Using Reconfigurable Hardware"

[Stephan Wong - EWI]

==============================================================
CALL FOR PAPERS

International Journal of Reconfigurable Computing

SPECIAL ISSUE ON
Accelerating (Deep) Neural Networks and Approximate Computing using Reconfigurable Hardware

Reconfigurable computing offers the promise of substantial performance and energy gains
over traditional architectures by customizing, even at runtime, the topology of the
underlying architecture to match the specific needs of a given application. Contemporary
adaptive and reconfigurable systems allow defining architectures with functional and
storage units that match the specific needs of a given computation in function, bit width,
and control structures. There are many applications that can take advantage of these
systems to address the challenge of increasing performance requirements with a low energy
budget. Recently, two application paradigms have been gaining an interest: (1) Deep Neural
Networks (DNNs) and other advanced Machine Learning techniques that, driven by the
availability of huge amounts of data, are becoming mainstream and (2)  Approximate
computing, a paradigm which, for the past years, has introduced quality as a new metric in
the design space of microprocessors, considering that many modern and meaningful
application domains, such as computer vision, multimedia processing, gaming, machine
learning, and data mining, are tolerant to some degree of imprecision. In light of this,
this special issue seeks high-quality contributions describing work that involves the use
of reconfigurable systems for DNNs and approximate computing, which may include: specific
accelerators and infrastructures; heterogeneous and multicore platforms; implementations
and comparisons to existing platforms; practical studies; languages, tools, frameworks and
design-flows; and emerging applications.

TOPICS
Potential topics include but are not limited to the following:
-- Reconfigurable architectures and/or FPGA implementations for DNNs, Machine Learning, or
   Approximate Computing
-- Heterogeneous and multicore platforms for DNNs and Machine Learning
-- Low-power reconfigurable architectures for DNNs and Machine Learning for training and/or
   inference
-- Embedded, SoC-based reconfigurable architectures for DNNs and Machine Learning for
   training and/or inference
-- Accelerators and communication infrastructures
-- Trade-offs between FPGAs, GPUs, and other accelerators for DNNs, Machine Learning, and
   Approximate Computing
-- Applications of DNNs, Machine Learning, and Approximate Computing (multimedia, automotive,
   robotic applications, bioinformatics, HPC applications, big data, security, etc.)
-- Domain-specific languages, tools, and programming frameworks for supporting DNNs,
   Machine Learning, and/or Approximate Computing
-- Design flows, codesign techniques, and high-level synthesis tools supporting DNNs,
   Machine Learning, and/or Approximate Computing
-- Autonomous and self-adaptive systems using DNNs and Machine Learning
-- Self-healing systems using DNNs and Machine Learning
-- Virtualization techniques
-- Industrial research and practice

IMPORTANT DATES AND SUBMISSION
Submission Deadline: Friday, 2 March 2018
Publication Date: July 2018

Authors can submit their manuscripts through the Manuscript Tracking System at
https://mts.hindawi.com/submit/journals/ijrc/anna/

Papers are published upon acceptance, regardless of the Special Issue publication date.

EDITORS
LEAD GUEST EDITOR
-- Ben Juurlink, Berlin University of Technology, Berlin, Germany
b.juurlink at tu-berlin.de

GUEST EDITORS
-- Georgios Keramidas, Think Silicon S.A./Technological Educational Institute of
   Western Greece, Patra, Greece
g.keramidas at think-silicon.com

-- Stephan Wong, Delft University of Technology, Delft, Netherlands
j.s.s.m.wong at tudelft.nl

-- Antonio C. S. Beck, Federal University of Rio Grande do Sul, Porto Alegre, Brazil
caco at inf.ufrgs.br

-- Chao Wang, University of Science and Technology of China, Suzhou, China
cswang at ustc.edu.cn

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mcs.anl.gov/mailman/private/hpc-announce/attachments/20171013/39c0964d/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 258652.pdf
Type: application/pdf
Size: 124893 bytes
Desc: 258652.pdf
URL: <https://lists.mcs.anl.gov/mailman/private/hpc-announce/attachments/20171013/39c0964d/attachment-0001.pdf>