Conference paper Open Access
A Novel Posit-based Fast Approximation of ELU Activation Function for Deep Neural Networks
Cococcioni, Marco; Rossi, Federico; Ruffaldi, Emanuele; Saponara, Sergio
Nowadays, real-time applications are exploiting DNNs more and more for computer vision and image recognition tasks. Such kind of applications are posing strict constraints in terms of both fast and efficient information representation and processing. New formats for representing real numbers have been proposed and among them the Posit format appears to be very promising, providing means to implement fast approximated version of widely used activation functions in DNNs. Moreover, information processing performance are continuously improved thanks to advanced vectorized SIMD (single-instruction multiple-data) processor architectures and instructions like ARM SVE (Scalable Vector Extension). This paper explores both approaches (Posit-based implementation of activation functions and vectorized SIMD processor architectures) to obtain faster DNNs. The two proposed techniques are able to speed up both DNN training and inference steps.
| Name | Size | |
|---|---|---|
|
ROSSI_C3_smart_computing_2020_su_3_pp.pdf
md5:c3961c8b727d556a9817d356a46728d3 |
191.4 kB | Download |
| All versions | This version | |
|---|---|---|
| Views | 80 | 80 |
| Downloads | 215 | 215 |
| Data volume | 41.2 MB | 41.2 MB |
| Unique views | 70 | 70 |
| Unique downloads | 207 | 207 |
- Publication date:
- September 22, 2020
- DOI:
-
- Published in:
- A Novel Posit-based Fast Approximation of ELU Activation Function for Deep Neural Networks: 2020 IEEE International Conference on Smart Computing (SMARTCOMP).
- Grants:
-
European Commission:
- EPI SGA1 - SGA1 (Specific Grant Agreement 1) OF THE EUROPEAN PROCESSOR INITIATIVE (EPI) (826647)
- Meeting:
- SMARTCOMP2020, Virtual (Online), 14/09/2022
- Related identifiers:
- Identical to
- 10.1109/SMARTCOMP50058.2020.00053 (Conference paper)
- License (for files):
- Creative Commons Attribution 1.0 Generic