G
Georgios Zervakis
Researcher at Karlsruhe Institute of Technology
Publications - 55
Citations - 708
Georgios Zervakis is an academic researcher from Karlsruhe Institute of Technology. The author has contributed to research in topics: Computer science & Artificial neural network. The author has an hindex of 10, co-authored 40 publications receiving 310 citations. Previous affiliations of Georgios Zervakis include National and Kapodistrian University of Athens & National Technical University of Athens.
Papers
More filters
Journal ArticleDOI
Design-Efficient Approximate Multiplication Circuits Through Partial Product Perforation
TL;DR: This paper proves in a mathematically rigorous manner that in partial product perforation, the imposed errors are bounded and predictable, depending only on the input distribution, in terms of power dissipation and error.
Journal ArticleDOI
Approximate Hybrid High Radix Encoding for Energy-Efficient Inexact Multipliers
TL;DR: This paper proposes an approximate hybrid high radix encoding for generating the partial products in signed multiplications that encodes the most significant bits with the accurate radix-4 encoding and the least significantbits with an approximate higher radix encode.
Journal ArticleDOI
Weight-Oriented Approximation for Energy-Efficient Neural Network Inference Accelerators
Zois-Gerasimos Tasoulas,Georgios Zervakis,Iraklis Anagnostopoulos,Hussam Amrouch,Jorg Henkel +4 more
TL;DR: A time-efficient automated framework for mapping the NN weights to the accuracy levels of the approximate reconfigurable accelerator that is able to satisfy tight accuracy loss thresholds, while significantly reducing energy consumption without any need for intensive NN retraining is proposed.
Proceedings ArticleDOI
Delta DICE: A Double Node Upset resilient latch
TL;DR: Simulation results demonstrated that in terms of power dissipation and propagation delay, the Delta DICE latch outperforms BISER-based latches that are SNU or DNU tolerant and provides DNU resilience at a small energy×delay penalty compared to other SNU tolerant cells.
Journal ArticleDOI
Walking through the Energy-Error Pareto Frontier of Approximate Multipliers
TL;DR: It is shown that design solutions configured through the proposed approach form the Pareto frontier of the energy-error space when considering direct quantitative comparisons with existing state-of-the-art design space.