Tech 'n' Test 'n' ML Research

Category: Conference

  • PASM Paper presented at HiPEAC2019

    PASM Paper presented at HiPEAC2019

    And that’s the PASM paper presented.

    Me presenting PASM at HiPEAC2019

    Lots of interest and very good questions from the audience and the session chair Luca Fanucci (Università di Pisa @Unipisa). Thoroughly enjoyed it. #HiPEAC19#hipeac2019

  • Come and see my talk at HiPEAC 2019

    I am presenting my second Ph.D. publication entitled “Low Complexity Multiply-Accumulate Units for Convolutional Neural Networks with Weight-Sharing” at HiPEAC 2019.

    PASM in a CNN Layer

    I will be presenting in Session 12 Programming Models, Neural Networks.

    If you can’t make the talk, I’ll also be presenting a poster of the paper in the Student Poster Session.

    Finally, if you can’t make the conference, you can still read my paper for free on the ACM TACO web site.

  • Second Ph.D. Paper Published by ACM TACO

    Second Ph.D. Paper Published by ACM TACO

    Dr. David Gregg and I have had my second Ph.D. paper published. Entitled “Low Complexity Multiply-Accumulate Units for Convolutional Neural Networks with Weight-Sharing”, the ACM Transactions on Architecture and Code Optimization published it in their August 2018 Journal. The ACM also published it on their online server, which can be viewed for free! It can also be found on the arXiv.org pre-print server.


    Simplified weight-shared convolution with PASM

    Here are more details of the publication:
    Title: Low Complexity Multiply-Accumulate Units for Convolutional Neural Networks with Weight-Sharing
    Published in:     ACM Transactions on Architecture and Code Optimization (TACO) (Volume: 15, Issue: 3, Article 31)
    Date of Publication: August 2018
    Print ISSN: 1544-3566
    DOI: 10.1145/3233300
    Publisher: ACM

    ACM TACO Bibtex (should you wish to cite our paper):

    @article{Garland:2018:LCM:3274266.3233300,
 author = {Garland, James and Gregg, David},
 title = {Low Complexity Multiply-Accumulate Units for Convolutional Neural Networks with Weight-Sharing},
 journal = {ACM Trans. Archit. Code Optim.},
 issue_date = {August 2018},
 volume = {15},
 number = {3},
 month = sep,
 year = {2018},
 issn = {1544-3566},
 pages = {31:1--31:24},
 articleno = {31},
 numpages = {24},
 url = {http://doi.acm.org/10.1145/3233300},
 doi = {10.1145/3233300},
 acmid = {3233300},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {ASIC, CNN, FPGA, arithmetic hardware circuits, multiply accumulate, power efficiency},
}

    Published in: arXiv.org. arXiv Bibtex:

    @article{DBLP:journals/corr/abs-1801-10219,
  author    = {James Garland and David Gregg},
  title     = {Low Complexity Multiply-Accumulate Units for Convolutional Neural Networks with Weight-Sharing},
  journal   = {CoRR},
  volume    = {abs/1801.10219},
  year      = {2018},
  url       = {http://arxiv.org/abs/1801.10219},
  archivePrefix = {arXiv},
  eprint    = {1801.10219},
  timestamp = {Mon, 13 Aug 2018 16:47:23 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1801-10219},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

    My complete list of publications can also be seen on my research publications page.

  • 13th Summer School on Advanced Computer Architecture and Compilation for High-Performance and Embedded Systems

    13th Summer School on Advanced Computer Architecture and Compilation for High-Performance and Embedded Systems

    Just a quick note from me to congratulate and thank the organisers and lecturers for what I feel was a very enjoyable, educational, informative and entertaining (Trunk Monkey [1] aside Prof. Huebner) set of courses and keynote talks. I’m still going over it all and digesting it more than a week later! And the photos [2] are great!

    The poster session was also an eye opener for me in that not only could I communicate my published paper as expected but also spread the word and hopefully get a few more citations for it 😉

    Both keynotes were excellent and I look forward to getting a (Google blessed) copy of the slides from Olivier. The career talk was also informative and useful so thanks to Xavier for arranging that.

    And last but not least, thank you Vicky for arranging for my family to stay at the hotel with me. They had a blast!

    [1] https://www.youtube.com/watch?v=wK47Xbmc-oQ&list=PLoxADsZ9YforxtjWODy3Kz_8JjLH-Nxp7
    [2] https://goo.gl/photos/KDyL5PFPc3HpQJtP6