Approximation-conscious IC testing
… and test cost, and based on distinct subsets of test … approximation-conscious multi-level IC
test flow, which classifies the output of the test process to be either: (1) a “good” defect-free IC, (…
test flow, which classifies the output of the test process to be either: (1) a “good” defect-free IC, (…
Highly-reliable approximate quadruple modular redundancy with approximation-aware voting
… Modern integrated circuits (ICs) are increasingly moving towards reduced feature sizes with
high integration density. However, this consequently results in reduced noise margins and …
high integration density. However, this consequently results in reduced noise margins and …
[PDF][PDF] Approximate tsv-based 3d stacked integrated circuits by inexact interconnects
MS Masadeh - … Journal of Advanced Computer Science and …, 2022 - researchgate.net
… Therefore, this work aims to increase the yield and reduce the test cost by accepting 3D-SICs
with defected interconnects as approximate 3D-SICs. This work considers 3DSICs, where …
with defected interconnects as approximate 3D-SICs. This work considers 3DSICs, where …
A machine learning-based error model for quality control of functional approximate computing
M Masadeh, O AlShorman… - … Conference on Data …, 2020 - ieeexplore.ieee.org
… The data was divided into two groups: the training-testing dataset and the validation dataset.
70% … The most accurate model was selected for testing stage based on 15% for the dataset. …
70% … The most accurate model was selected for testing stage based on 15% for the dataset. …
Machine learning-based self-compensating approximate computing
Dedicated hardware accelerators are suitable for parallel computational tasks. Moreover, they
have the tendency to accept inexact results. These hardware accelerators are extensively …
have the tendency to accept inexact results. These hardware accelerators are extensively …
Machine-learning-based self-tunable design of approximate computing
… -testing … the testing stage based on 15% for the data set. Finally, we validated the model
using the remaining 15% for the data set. Thus, we ensure that no training data leak into the test …
using the remaining 15% for the data set. Thus, we ensure that no training data leak into the test …
[PDF][PDF] Machine Learning-Based Self-Compensating Approximate Computing for Autonomous Systems
Dedicated hardware accelerators are suitable for parallel computational tasks. Moreover, they
have the tendency to accept inexact results. These hardware accelerators are extensively …
have the tendency to accept inexact results. These hardware accelerators are extensively …
[PDF][PDF] An efficient machine learning-based COVID-19 identification utilizing chest X-ray images
M Masadeh, A Masadeh, O Alshorman… - … Journal of Artificial …, 2022 - researchgate.net
… Thus, our model can be realized globally for efficient monitoring and prioritization of testing
for the … For parameter tuning, we used 464 images while performed final model testing by the …
for the … For parameter tuning, we used 464 images while performed final model testing by the …
[BOOK][B] Biosociology of dominance and deference
A Mazur - 2005 - books.google.com
… The criterion of breeding fertile offspring cannot be tested on extinct species, so … As an
example, we might compare DNA from lions, tigers, domestic cats, and dogs. (Dogs, …
example, we might compare DNA from lions, tigers, domestic cats, and dogs. (Dogs, …
[BOOK][B] In Search of Interlanguage; a Study of Second-language Acquisition of German Syntax
CMB Garrett - 1982 - search.proquest.com
… ln the first category we include sociol inguist ic and pragmatic determinants such as …
The des i gn of this study therefore încluded two tests for deci di ng which forms in learners" …
The des i gn of this study therefore încluded two tests for deci di ng which forms in learners" …