most citedautoAx: An Automatic Design Space Exploration and Circuit Building Methodology utilizing Libraries of Approximate Components

74 citations · 102 across the 3 of their papers we have counts for

collaborators

8 papers

cs.NE202016 cited

Semantically-Oriented Mutation Operator in Cartesian Genetic Programming for Evolutionary Circuit Design

David Hodan, Vojtech Mrazek, Zdenek Vasicek

Despite many successful applications, Cartesian Genetic Programming (CGP) suffers from limited scalability, especially when used for evolutionary circuit design. Considering the mu…

cs.AR202012 cited

Using Libraries of Approximate Circuits in Design of Hardware Accelerators of Deep Neural Networks

Vojtech Mrazek, Lukas Sekanina, Zdenek Vasicek

Approximate circuits have been developed to provide good tradeoffs between power consumption and quality of service in error resilient applications such as hardware accelerators of…

cs.NE2020

Adaptive Verifiability-Driven Strategy for Evolutionary Approximation of Arithmetic Circuits

Milan Ceska, Jiri Matyas, Vojtech Mrazek +3

We present a novel approach for designing complex approximate arithmetic circuits that trade correctness for power consumption and play important role in many energy-aware applicat…

cs.DC2020

TFApprox: Towards a Fast Emulation of DNN Approximate Hardware Accelerators on GPU

Filip Vaverka, Vojtech Mrazek, Zdenek Vasicek +1

Energy efficiency of hardware accelerators of deep neural networks (DNN) can be improved by introducing approximate arithmetic circuits. In order to quantify the error introduced b…

cs.LG2019

ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations

Alberto Marchisio, Vojtech Mrazek, Muhammad Abudllah Hanif +1

Recent advances in Capsule Networks (CapsNets) have shown their superior learning capability, compared to the traditional Convolutional Neural Networks (CNNs). However, the extreme…

cs.NE2019

ALWANN: Automatic Layer-Wise Approximation of Deep Neural Network Accelerators without Retraining

Vojtech Mrazek, Zdenek Vasicek, Lukas Sekanina +2

The state-of-the-art approaches employ approximate computing to reduce the energy consumption of DNN hardware. Approximate DNNs then require extensive retraining afterwards to reco…