Details
| Original language | English |
|---|---|
| Title of host publication | SPLC 2019 |
| Subtitle of host publication | 23rd International Systems and Software Product Line Conference |
| Editors | Thorsten Berger, Philippe Collet, Laurence Duchien, Thomas Fogdal, Patrick Heymans, Timo Kehrer, Jabier Martinez, Raul Mazo, Leticia Montalvillo, Camille Salinesi, Xhevahire Ternava, Thomas Thum, Tewfik Ziadi |
| Place of Publication | New York |
| Publisher | Association for Computing Machinery (ACM) |
| Pages | 72-77 |
| Number of pages | 6 |
| ISBN (electronic) | 9781450371384 |
| Publication status | Published - 9 Sept 2019 |
| Event | 23rd International Software Product Line Conference, SPLC 2019 - Paris, France Duration: 9 Sept 2019 → 13 Sept 2019 |
Publication series
| Name | ACM International Conference Proceeding Series (ICPS) |
|---|---|
| Volume | A |
Abstract
Deep Neural Networks (DNNs) are increasingly being used as a machine learning solution thanks to the complexity of their architecture and hyperparameters-weights. A drawback is the excessive demand for massive computational power during the training process. Not only as a whole but parts of neural networks can also be in charge of certain functionalities. We present a novel challenge in an intersection between machine learning and variability management communities to reuse modules of DNNs without further training. Let us assume that we are given a DNN for image processing that recognizes cats and dogs. By extracting a part of the network, without additional training a new DNN should be divisible with the functionality of recognizing only cats. Existing research in variability management can offer a foundation for a product line of DNNs composing the reusable functionalities. An ideal solution can be evaluated based on its speed, granularity of determined functionalities, and the support for adding variability to the network. The challenge is decomposed in three subchallenges: feature extraction, feature abstraction, and the implementation of a product line of DNNs.
Keywords
- Deep neural networks, Machine learning, Software product lines, Transfer learning, Variability
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Human-Computer Interaction
- Computer Science(all)
- Computer Vision and Pattern Recognition
- Computer Science(all)
- Computer Networks and Communications
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
SPLC 2019 : 23rd International Systems and Software Product Line Conference. ed. / Thorsten Berger; Philippe Collet; Laurence Duchien; Thomas Fogdal; Patrick Heymans; Timo Kehrer; Jabier Martinez; Raul Mazo; Leticia Montalvillo; Camille Salinesi; Xhevahire Ternava; Thomas Thum; Tewfik Ziadi. New York: Association for Computing Machinery (ACM), 2019. p. 72-77 (ACM International Conference Proceeding Series (ICPS); Vol. A).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Applying product line engineering concepts to deep neural networks
AU - Ghofrani, Javad
AU - Kozegar, Ehsan
AU - Fehlhaber, Anna Lena
AU - Soorati, Mohammad Divband
PY - 2019/9/9
Y1 - 2019/9/9
N2 - Deep Neural Networks (DNNs) are increasingly being used as a machine learning solution thanks to the complexity of their architecture and hyperparameters-weights. A drawback is the excessive demand for massive computational power during the training process. Not only as a whole but parts of neural networks can also be in charge of certain functionalities. We present a novel challenge in an intersection between machine learning and variability management communities to reuse modules of DNNs without further training. Let us assume that we are given a DNN for image processing that recognizes cats and dogs. By extracting a part of the network, without additional training a new DNN should be divisible with the functionality of recognizing only cats. Existing research in variability management can offer a foundation for a product line of DNNs composing the reusable functionalities. An ideal solution can be evaluated based on its speed, granularity of determined functionalities, and the support for adding variability to the network. The challenge is decomposed in three subchallenges: feature extraction, feature abstraction, and the implementation of a product line of DNNs.
AB - Deep Neural Networks (DNNs) are increasingly being used as a machine learning solution thanks to the complexity of their architecture and hyperparameters-weights. A drawback is the excessive demand for massive computational power during the training process. Not only as a whole but parts of neural networks can also be in charge of certain functionalities. We present a novel challenge in an intersection between machine learning and variability management communities to reuse modules of DNNs without further training. Let us assume that we are given a DNN for image processing that recognizes cats and dogs. By extracting a part of the network, without additional training a new DNN should be divisible with the functionality of recognizing only cats. Existing research in variability management can offer a foundation for a product line of DNNs composing the reusable functionalities. An ideal solution can be evaluated based on its speed, granularity of determined functionalities, and the support for adding variability to the network. The challenge is decomposed in three subchallenges: feature extraction, feature abstraction, and the implementation of a product line of DNNs.
KW - Deep neural networks
KW - Machine learning
KW - Software product lines
KW - Transfer learning
KW - Variability
UR - http://www.scopus.com/inward/record.url?scp=85123041875&partnerID=8YFLogxK
U2 - 10.1145/3336294.3336321
DO - 10.1145/3336294.3336321
M3 - Conference contribution
AN - SCOPUS:85123041875
T3 - ACM International Conference Proceeding Series (ICPS)
SP - 72
EP - 77
BT - SPLC 2019
A2 - Berger, Thorsten
A2 - Collet, Philippe
A2 - Duchien, Laurence
A2 - Fogdal, Thomas
A2 - Heymans, Patrick
A2 - Kehrer, Timo
A2 - Martinez, Jabier
A2 - Mazo, Raul
A2 - Montalvillo, Leticia
A2 - Salinesi, Camille
A2 - Ternava, Xhevahire
A2 - Thum, Thomas
A2 - Ziadi, Tewfik
PB - Association for Computing Machinery (ACM)
CY - New York
T2 - 23rd International Software Product Line Conference, SPLC 2019
Y2 - 9 September 2019 through 13 September 2019
ER -