77 citations · 104 across the 7 of their papers we have counts for
7 papers
Logic Shrinkage: Learned FPGA Netlist Sparsity for Efficient Neural Network Inference
Erwei Wang, James J. Davis, Georgios-Ilias Stavrou +3
FPGA-specific DNN architectures using the native LUTs as independently trainable inference operators have been shown to achieve favorable area-accuracy and energy-accuracy tradeoff…
Enabling Binary Neural Network Training on the Edge
Erwei Wang, James J. Davis, Daniele Moro +6
The ever-growing computational demands of increasingly complex machine learning models frequently necessitate the use of powerful cloud-based infrastructure for their training. Bin…
Digit Stability Inference for Iterative Methods Using Redundant Number Representation
He Li, Ian McInerney, James J. Davis +1
In our recent work on iterative computation in hardware, we showed that arbitrary-precision solvers can perform more favorably than their traditional arithmetic equivalents when th…
LUTNet: Learning FPGA Configurations for Highly Efficient Neural Network Inference
Erwei Wang, James J. Davis, Peter Y. K. Cheung +1
Research has shown that deep neural networks contain significant redundancy, and thus that high classification accuracy can be achieved even when weights and activations are quanti…
ARCHITECT: Arbitrary-precision Hardware with Digit Elision for Efficient Iterative Compute
He Li, James J. Davis, John Wickerson +1
Many algorithms feature an iterative loop that converges to the result of interest. The numerical operations in such algorithms are generally implemented using finite-precision ari…
LUTNet: Rethinking Inference in FPGA Soft Logic
Erwei Wang, James J. Davis, Peter Y. K. Cheung +1
Research has shown that deep neural networks contain significant redundancy, and that high classification accuracies can be achieved even when weights and activations are quantised…
Deep Neural Network Approximation for Custom Hardware: Where We've Been, Where We're Going
Erwei Wang, James J. Davis, Ruizhe Zhao +5
Deep neural networks have proven to be particularly effective in visual and audio recognition tasks. Existing models tend to be computationally expensive and memory intensive, howe…