TY - GEN
T1 - Improving Residue-Level Sparsity in RNS-based Neural Network Hardware Accelerators via Regularization
AU - Kavvousanos, E.
AU - Sakellariou, V.
AU - Kouretas, I.
AU - Paliouras, V.
AU - Stouraitis, T.
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Residue Number System (RNS) has recently attracted interest for the hardware implementation of inference in machine-learning systems as it provides promising trade-offs in the area, time, and power dissipation space. In this paper we introduce a technique that utilizes regularization during training, and increases the percentage of residues which are zero, when the parameters of an artificial neural network (ANN) are expressed in an RNS. The proposed technique can also be used as a post-processing stage, allowing the optimization of pre-trained models for RNS implementation. By increasing the number of residues being zero, i.e., residue-level sparsity, the proposed technique facilitates new hardware architectures for RNS-based inference, allowing new trade-offs and improving performance over prior art without practically compromising accuracy. The introduced method increases residue sparsity by a factor of 4× to 6× in certain cases.
AB - Residue Number System (RNS) has recently attracted interest for the hardware implementation of inference in machine-learning systems as it provides promising trade-offs in the area, time, and power dissipation space. In this paper we introduce a technique that utilizes regularization during training, and increases the percentage of residues which are zero, when the parameters of an artificial neural network (ANN) are expressed in an RNS. The proposed technique can also be used as a post-processing stage, allowing the optimization of pre-trained models for RNS implementation. By increasing the number of residues being zero, i.e., residue-level sparsity, the proposed technique facilitates new hardware architectures for RNS-based inference, allowing new trade-offs and improving performance over prior art without practically compromising accuracy. The introduced method increases residue sparsity by a factor of 4× to 6× in certain cases.
KW - hardware acceleration
KW - neural networks
KW - residue number system
KW - sparsity
UR - https://www.scopus.com/pages/publications/85189299749
U2 - 10.1109/ARITH58626.2023.00020
DO - 10.1109/ARITH58626.2023.00020
M3 - Conference contribution
AN - SCOPUS:85189299749
T3 - Proceedings - Symposium on Computer Arithmetic
SP - 102
EP - 109
BT - Proceedings - 2023 IEEE 30th Symposium on Computer Arithmetic, ARITH 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 30th IEEE Symposium on Computer Arithmetic, ARITH 2023
Y2 - 4 September 2023 through 6 September 2023
ER -