Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks-Reference-Cited by-同舟云学术

Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks

Published:2019-03-13 Issue: Volume: Page:
ISSN:
Container-title:Proceedings of the Conference for Next Generation Arithmetic 2019
language:
Short-container-title:

Author:

Carmichael Zachariah¹,Langroudi Hamed F.¹,Khazanov Char¹,Lillie Jeffrey¹,Gustafson John L.²,Kudithipudi Dhireesha¹

Affiliation:

1. Neuromorphic AI Lab, Rochester Institute of Technology, Rochester, NY

2. National University of Singapore, Singapore

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3316279.3316282

Reference33 articles.

1. Bengio Y. 2013. Deep Learning of Representations: Looking Forward. In Statistical Language and Speech Processing Adrian-Horia Dediu Carlos Martín-Vide Ruslan Mitkov and Bianca Truthe (Eds.). Springer Berlin Heidelberg Berlin Heidelberg 1--37. 10.1007/978-3-642-39593-2_1 Bengio Y. 2013. Deep Learning of Representations: Looking Forward. In Statistical Language and Speech Processing Adrian-Horia Dediu Carlos Martín-Vide Ruslan Mitkov and Bianca Truthe (Eds.). Springer Berlin Heidelberg Berlin Heidelberg 1--37. 10.1007/978-3-642-39593-2_1

2. A matrix-multiply unit for posits in reconfigurable logic leveraging (open)CAPI

3. Serving DNNs in Real Time at Datacenter Scale with Project Brainwave

Cited by 49 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. VitBit: Enhancing Embedded GPU Performance for AI Workloads through Register Operand Packing;Proceedings of the 53rd International Conference on Parallel Processing;2024-08-12

2. Low-Precision Mixed-Computation Models for Inference on Edge;IEEE Transactions on Very Large Scale Integration (VLSI) Systems;2024-08

3. Design Exploration of Fault-Tolerant Deep Neural Networks Using Posit Number Representation System;IEEE Transactions on Very Large Scale Integration (VLSI) Systems;2024-07

4. Implementation of Neural Network using Posit Arithmetic Circuits;2023 4th International Conference on Intelligent Technologies (CONIT);2024-06-21

5. ADEPNET: A Dynamic-Precision Efficient Posit Multiplier for Neural Networks;IEEE Access;2024