EPI Consortium members published “FPPU: Design and Implementation of a Pipelined Full Posit Processing Unit” in arXiv.
Here is a link to an open-access version of the article: https://arxiv.org/abs/2308.03425
DOI: https://doi.org/10.48550/arXiv.2308.03425
EPI Consortium members published “Experimental Results of Vectorized Posit-Based DNNs on a Real ARM SVE High Performance Computing Machine” in ZENODO.
Here is a link to an open-access version of the article: https://zenodo.org/record/7128765#.ZAHSvXbMIuU.
DOI: https://zenodo.org/badge/DOI/10.5281/zenodo.7128765.svg.
EPI Consortium members published “Small reals representations for Deep Learning at the edge: a comparison” at the 2022 edition of the Conference on Next Generation Arithmetic (CoNGA’22).
Here you can find a link to an open access version of the article: https://www.researchgate.net/publication/358884612_Small_reals_representations_for_Deep_Learning_at_the_edge_a_comparison.
DOI: https://doi.org/10.1007/978-3-031-09779-9_8.
EPI Consortium members published “A Lightweight Posit Processing Unit for RISC-V Processors in Deep Neural Network Applications” in IEEE Transactions on Emerging Topics in Computing.
Here you can find a link to an open access version of the article: https://ieeexplore.ieee.org/document/9583876.
DOI: https://doi.ieeecomputersociety.org/10.1109/TETC.2021.3120538.
EPI Consortium members published “Vectorizing posit operations on RISC-V for faster deep neural networks: experiments and comparison with ARM SVE” in Neural Computing & Applications.
Here you can find a link to an open access version of the article:
https://link.springer.com/article/10.1007/s00521-021-05814-0
DOI: https://doi.org/10.1007/s00521-021-05814-0
EPI Consortium members published “Novel Arithmetics in Deep Neural Networks Signal Processing for Autonomous Driving: Challenges and Opportunities” in the IEEE Signal Processing Magazine.
Here you can find a link to an open access version of the article:
https://zenodo.org/record/4564326
DOI: https://doi.org/10.1109/MSP.2020.2988436
EPI Consortium members published “A Novel Posit-based Fast Approximation of ELU Activation Function for Deep Neural Networks” in I2020 IEEE International Conference on Smart Computing (SMARTCOMP).
Here you can find a link to an open access version of the article:
https://zenodo.org/record/4042854#.X6hQDYhKiUk
DOI: https://doi.org/10.1109/SMARTCOMP50058.2020.00053
Our colleagues from UNIPI presented at SMARTCOMP, with a paper titled: A Novel Posit-based Fast Approximation of ELU Activation Function for Deep Neural Networks. The poster from the conference is available here.
EPI Consortium members published “Fast deep neural networks for image processing using posits and ARM scalable vector extension” in Journal of Real-Time Image Processing volume 17, pages759–771(2020).
Here you can find a link to an open access version of the article: https://link.springer.com/article/10.1007/s11554-020-00984-x.
DOI: https://doi.org/10.1007/s11554-020-00984-x
EPI Consortium members published “AFast Approximations of Activation Functions in Deep Neural Networks when using Posit Arithmetic” in the Sensors 2020, 20(05), 1515.
Here you can find a link to an open access version of the article: