Skip to main content

Showing 1–2 of 2 results for author: Belk, J

Searching in archive cs. Search in all archives.
.
  1. arXiv:2209.01346  [pdf, other

    cs.DC cs.AI cs.AR cs.NI cs.PF

    HammingMesh: A Network Topology for Large-Scale Deep Learning

    Authors: Torsten Hoefler, Tommaso Bonato, Daniele De Sensi, Salvatore Di Girolamo, Shigang Li, Marco Heddes, Jon Belk, Deepak Goel, Miguel Castro, Steve Scott

    Abstract: Numerous microarchitectural optimizations unlocked tremendous processing power for deep neural networks that in turn fueled the AI revolution. With the exhaustion of such optimizations, the growth of modern AI is now gated by the performance of training systems, especially their data movement. Instead of focusing on single accelerators, we investigate data-movement characteristics of large-scale t… ▽ More

    Submitted 21 October, 2022; v1 submitted 3 September, 2022; originally announced September 2022.

    Comments: published at ACM/IEEE Supercomputing (SC22)

  2. arXiv:1405.0982  [pdf, other

    math.GR cs.FL math.DS math.LO

    Some undecidability results for asynchronous transducers and the Brin-Thompson group 2V

    Authors: James Belk, Collin Bleak

    Abstract: Using a result of Kari and Ollinger, we prove that the torsion problem for elements of the Brin-Thompson group 2V is undecidable. As a result, we show that there does not exist an algorithm to determine whether an element of the rational group R of Grigorchuk, Nekrashevich, and Sushchanskii has finite order. A modification of the construction gives other undecidability results about the dynamics o… ▽ More

    Submitted 5 May, 2014; originally announced May 2014.

    Comments: 16 pages, 3 figures

    MSC Class: 20F10 (Primary) 68Q45; 37B99; 03B25 (Secondary)

    Journal ref: Trans. Amer. Math. Soc. 369 (2017), 3157-3172