Nonconventional computer arithmetic circuits, systems and applications

L Sousa - IEEE Circuits and Systems Magazine, 2021 - ieeexplore.ieee.org
Arithmetic plays a major role in a computer? s performance and efficiency. Building new
computing platforms supported by the traditional binary arithmetic and silicon-based …

Tuning Strassen's matrix multiplication for memory efficiency

M Thottethodi, S Chatterjee… - SC'98: Proceedings of …, 1998 - ieeexplore.ieee.org
Strassen's algorithm for matrix multiplication gains its lower arithmetic complexity at the
expense of reduced locality of reference, which makes it challenging to implement the …

Stochastic processors on FPGAs to compute sensor data towards fault-tolerant IoT systems

RP Duarte, HC Neto - 2018 IEEE Conference on Dependable …, 2018 - ieeexplore.ieee.org
The continuous increase in the amounts of data received at the edges of the Grid is pushing
the pre-computation of sensor data at the IoT device before communicating it over the …

Dither computing: a hybrid deterministic-stochastic computing framework

CW Wu - 2021 IEEE 28th Symposium on Computer Arithmetic …, 2021 - ieeexplore.ieee.org
Stochastic computing has a long history as an alternative method of performing arithmetic on
a computer. While it can be considered an unbiased estimator of real numbers, it has a …

Xtokaxtikox: A stochastic computing-based autonomous cyber-physical system

RP Duarte, H Neto, M Véstias - 2016 IEEE International …, 2016 - ieeexplore.ieee.org
This paper presents XtokaxtikoX, a fully autonomous cyber-physical system employing only
stochastic arithmetic to perform computations on its data-path. Traditional implementations …

[PDF][PDF] Rapid prototyping of approximate signal processing using stochastic processors on FPGAs

RP Duarte, H Neto - … Conference on Design and Architectures for …, 2018 - tecnico.ulisboa.pt
In the context of IoT, silicon devices are employed to perform intense signal processing
computations. They tend to operate under heavy constraints for power and area, traditional …

[PDF][PDF] Event driven fully digital neural net

A Venzo - 2020 - webthesis.biblio.polito.it
Machine learning and neural net use is increasing for many tasks, which require high
computational parallelism and but also high energy efficiency and low power consumption …

Stochastic Theater: Stochastic Datapath Generation Framework for Fault-Tolerant IoT Sensors

RP Duarte, M Véstias, C Carvalho… - i-ETC: ISEL Academic …, 2018 - 192.68.221.50
Stochastic Computing has emerged as a competitive computing paradigm that produces fast
and simple implementations of arithmetic operations, while offering high levels of …