activity
20152023
most citedLow-memory GEMM-based convolution algorithms for deep neural networks

44 citations · 167 across the 26 of their papers we have counts for

collaborators
Showing math.NAShow all

2 papers · 1 filter

math.NA2021★ 15 cited

Low precision logarithmic number systems: Beyond base-2

Syed Asad Alam, James Garland, David Gregg

Logarithmic number systems (LNS) are used to represent real numbers in many applications using a constant base raised to a fixed-point exponent making its distribution exponential.…

math.NA2018

Error Analysis and Improving the Accuracy of Winograd Convolution for Deep Neural Networks

Barbara Barabasz, Andrew Anderson, Kirk M. Soodhalter +1

Popular deep neural networks (DNNs) spend the majority of their execution time computing convolutions. The Winograd family of algorithms can greatly reduce the number of arithmetic…