Residue Number

Residue number systems (RNS) are mathematical representations of numbers that offer advantages in computational efficiency, particularly for applications requiring high-speed arithmetic like digital signal processing and deep learning. Current research focuses on improving RNS performance through hybrid systems combining RNS with redundant number systems or high-dimensional vector representations, and on applying RNS to enhance the accuracy and energy efficiency of analog deep neural network accelerators. These advancements hold significant promise for accelerating computations in various fields, from improving the speed and power consumption of electronic devices to enabling more efficient analysis of large biological datasets.

Papers

August 10, 2024

Enhancing Computational Efficiency in Intensive Domains via Redundant Residue Number Systems
Soudabeh Mousavi, Dara Rahmati, Saeid Gorgin, Jeong-A Lee
Computational Efficiency Numerical Method Challenging Domain Residue Number

May 3, 2024

Protein binding affinity prediction under multiple substitutions applying eGNNs on Residue and Atomic graphs combined with Language model information: eGRAL
Arturo Fiorellini-Bernardis, Sebastien Boyer, Christoph Brunken, Bakary Diallo, Karim Beguir, Nicolas Lopez-Carranza, Oliver Bent
High Throughput Affinity Prediction Protein Data Back Substitution Affinity Loss Residue Number

November 8, 2023

Computing with Residue Numbers in High-Dimensional Representation
Christopher J. Kymn, Denis Kleyko, E. Paxon Frady, Connor Bybee, Pentti Kanerva, Friedrich T. Sommer, Bruno A. Olshausen
Hyperdimensional Computing Digital Computing High Dimensional Representation Residue Number

June 15, 2023

Leveraging Residue Number System for Designing High-Precision Analog Deep Neural Network Accelerators
Cansu Demirkiran, Rashmi Agrawal, Vijay Janapa Reddi, Darius Bunandar, Ajay Joshi
Low Precision Analog Neural Network Analog Accelerator Residue Number

April 3, 2023

Classification of integers based on residue classes via modern deep learning algorithms
Da Wu, Jingye Yang, Mian Umair Ahsan, Kai Wang
Machine Learning Machine Learning Model Classification Code Neural Architecture Deep Learning Algorithm Feature Space Integer Arithmetic Prime Number Residue Number

November 12, 2022

Integrating Transformer and Autoencoder Techniques with Spectral Graph Algorithms for the Prediction of Scarcely Labeled Molecular Data
Nicole Hayes, Ekaterina Merkurjev, Guo-Wei Wei
Transfer Learning Transformer Based Human Prediction Graph Based Numerical Experiment Autoencoder Based Molecular Data Spectral Graph Residue Number

Residue Number

Papers

Enhancing Computational Efficiency in Intensive Domains via Redundant Residue Number Systems

Protein binding affinity prediction under multiple substitutions applying eGNNs on Residue and Atomic graphs combined with Language model information: eGRAL

Computing with Residue Numbers in High-Dimensional Representation

Leveraging Residue Number System for Designing High-Precision Analog Deep Neural Network Accelerators

Classification of integers based on residue classes via modern deep learning algorithms

Integrating Transformer and Autoencoder Techniques with Spectral Graph Algorithms for the Prediction of Scarcely Labeled Molecular Data