Abstract | ||
---|---|---|
Learning on chip (LOC) is a challenging problem in which an embedded system learns a model and uses it to process and classify unknown data, while adapting to new observations or classes. It may require intensive computations and complex hardware implementations to adapt to new data. We address this issue by introducing an incremental learning method based on the combination of a pre-trained Convolutional Neural Network (CNN) and majority votes, using Product Quantizing (PQ) as a bridge between them. We detail a hardware implementation of the proposed method (validated on a FPGA target) using limited hardware resources while providing substantial processing acceleration compared to a CPU counterpart. |
Year | Venue | Keywords |
---|---|---|
2017 | IEEE Global Conference on Signal and Information Processing | transfer learning,incremental learning,learning on chip,convolutional neural network,FPGA |
Field | DocType | ISSN |
Hardware implementations,Computer science,Convolutional neural network,Transfer of learning,Incremental learning,Field-programmable gate array,Acceleration,Quantization (signal processing),Computer engineering,Computation | Conference | 2376-4066 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ghouthi Boukli Hacene | 1 | 1 | 1.04 |
Vincent Gripon | 2 | 210 | 27.16 |
Nicolas Farrugia | 3 | 21 | 4.16 |
Matthieu Arzel | 4 | 69 | 15.10 |
Michel Jézéquel | 5 | 769 | 84.23 |