A blueprint for precise and fault-tolerant analog neural networks
Analog computing has reemerged as a promising avenue for accelerating deep neural
networks (DNNs) to overcome the scalability challenges posed by traditional digital …
networks (DNNs) to overcome the scalability challenges posed by traditional digital …
Rnsnet: In-memory neural network acceleration using residue number system
We live in a world where technological advances are continually creating more data than
what we can deal with. Machine learning algorithms, in particular Deep Neural Networks …
what we can deal with. Machine learning algorithms, in particular Deep Neural Networks …
RNS-To-Binary Converter for a New Three-Moduli Set
PVA Mohan - IEEE Transactions on Circuits and Systems II …, 2007 - ieeexplore.ieee.org
In this brief, the design of residue number system (RNS) to binary converters for a new
powers-of-two related three-moduli set {2 n+ 1-1, 2 n, 2 n-1} is considered. This moduli set …
powers-of-two related three-moduli set {2 n+ 1-1, 2 n, 2 n-1} is considered. This moduli set …
A residue-to-binary converter for a new five-moduli set
The efficiency of the residue number system (RNS) depends not only on the residue-to-
binary converters but also the operand sizes and the modulus in each residue channel. Due …
binary converters but also the operand sizes and the modulus in each residue channel. Due …
Modular multiplication and base extensions in residue number systems
We present a new RNS modular multiplication for very large operands. The algorithm is
based on Montgomery's (1985) method adapted to residue arithmetic. By choosing the …
based on Montgomery's (1985) method adapted to residue arithmetic. By choosing the …
RNS-to-Binary Converters for Two Four-Moduli Sets and
PVA Mohan, AB Premkumar - IEEE Transactions on Circuits …, 2007 - ieeexplore.ieee.org
In this paper, reverse converters for two recently proposed four-moduli sets {2 n-1, 2 n, 2 n+
1, 2 n+ 1-1} and {2 n-1, 2 n, 2 n+ 1, 2 n+ 1+ 1} are described. The reverse conversion in the …
1, 2 n+ 1-1} and {2 n-1, 2 n, 2 n+ 1, 2 n+ 1+ 1} are described. The reverse conversion in the …
High-speed and reduced-area modular adder structures for RNS
AA Hiasat - IEEE Transactions on Computers, 2002 - ieeexplore.ieee.org
A modular adder is a very instrumental arithmetic component in implementing online residue-
based computations for many digital signal processing applications. It is also a basic …
based computations for many digital signal processing applications. It is also a basic …
New Chinese remainder theorems
Y Wang - Conference Record of Thirty-Second Asilomar …, 1998 - ieeexplore.ieee.org
The residue-to-binary conversion is the crucial step for residue arithmetic. The traditional
methods are the Chinese remainder theorem (CRT) and the mixed radix conversion. This …
methods are the Chinese remainder theorem (CRT) and the mixed radix conversion. This …
A study of the residue-to-binary converters for the three-moduli sets
In this paper, a detailed study on the four three-moduli sets reported in the literature is
carried out from the point of view of the hardware complexity and speed of their residue-to …
carried out from the point of view of the hardware complexity and speed of their residue-to …
A high-speed residue-to-binary converter for three-moduli (2/sup k/, 2/sup k/-1, 2/sup k-1/-1) RNS and a scheme for its VLSI implementation
In this paper, a high-speed residue-to-binary converter for the moduli set (2/sup k/, 2/sup k/-
1, 2/sup k-1/-1) is proposed. Compared to the previous converter based on this moduli set …
1, 2/sup k-1/-1) is proposed. Compared to the previous converter based on this moduli set …