Skip to main content

2018 | OriginalPaper | Buchkapitel

Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weights

verfasst von : Arun Mallya, Dillon Davis, Svetlana Lazebnik

Erschienen in: Computer Vision – ECCV 2018

Verlag: Springer International Publishing

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

This work presents a method for adapting a single, fixed deep neural network to multiple tasks without affecting performance on already learned tasks. By building upon ideas from network quantization and pruning, we learn binary masks that “piggyback” on an existing network, or are applied to unmodified weights of that network to provide good performance on a new task. These masks are learned in an end-to-end differentiable fashion, and incur a low overhead of 1 bit per network parameter, per task. Even though the underlying network is fixed, the ability to mask individual weights allows for the learning of a large number of filters. We show performance comparable to dedicated fine-tuned networks for a variety of classification tasks, including those with large domain shifts from the initial task (ImageNet), and a variety of network architectures. Our performance is agnostic to task ordering and we do not suffer from catastrophic forgetting or competition between tasks.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Fußnoten
1
This is lower than the 63.6 mIOU obtained by [32] owing to differences in the Caffe and PyTorch VGG-16 initializations, as documented at https://​goo.​gl/​quvmm2.
 
Literatur
1.
Zurück zum Zitat Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. CoRR abs/1409.1556 (2014) Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. CoRR abs/1409.1556 (2014)
2.
Zurück zum Zitat Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. IJCV (2015) Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. IJCV (2015)
3.
Zurück zum Zitat French, R.M.: Catastrophic forgetting in connectionist networks. Trends Cogn. Sci. 3(4), 128–135 (1999)CrossRef French, R.M.: Catastrophic forgetting in connectionist networks. Trends Cogn. Sci. 3(4), 128–135 (1999)CrossRef
4.
Zurück zum Zitat Kirkpatrick, J., et al.: Overcoming catastrophic forgetting in neural networks. In: PNAS (2017) Kirkpatrick, J., et al.: Overcoming catastrophic forgetting in neural networks. In: PNAS (2017)
5.
Zurück zum Zitat Rannen, A., Aljundi, R., Blaschko, M.B., Tuytelaars, T.: Encoder based lifelong learning. In: ICCV (2017) Rannen, A., Aljundi, R., Blaschko, M.B., Tuytelaars, T.: Encoder based lifelong learning. In: ICCV (2017)
7.
8.
Zurück zum Zitat Wah, C., Branson, S., Welinder, P., Perona, P., Belongie, S.: The Caltech-UCSD Birds-200-2011 Dataset. Technical report CNS-TR-2011-001, California Institute of Technology (2011) Wah, C., Branson, S., Welinder, P., Perona, P., Belongie, S.: The Caltech-UCSD Birds-200-2011 Dataset. Technical report CNS-TR-2011-001, California Institute of Technology (2011)
9.
Zurück zum Zitat Krause, J., Stark, M., Deng, J., Fei-Fei, L.: 3D object representations for fine-grained categorization. In: CVPRW (2013) Krause, J., Stark, M., Deng, J., Fei-Fei, L.: 3D object representations for fine-grained categorization. In: CVPRW (2013)
10.
Zurück zum Zitat Nilsback, M.E., Zisserman, A.: Automated flower classification over a large number of classes. In: ICCVGIP (2008) Nilsback, M.E., Zisserman, A.: Automated flower classification over a large number of classes. In: ICCVGIP (2008)
11.
Zurück zum Zitat Saleh, B., Elgammal, A.: Large-scale classification of fine-art paintings: Learning the right metric on the right feature. In: ICDMW (2015) Saleh, B., Elgammal, A.: Large-scale classification of fine-art paintings: Learning the right metric on the right feature. In: ICDMW (2015)
12.
Zurück zum Zitat Eitz, M., Hays, J., Alexa, M.: How do humans sketch objects? In: SIGGRAPH (2012) Eitz, M., Hays, J., Alexa, M.: How do humans sketch objects? In: SIGGRAPH (2012)
13.
Zurück zum Zitat He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016) He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)
14.
Zurück zum Zitat Zagoruyko, S., Komodakis, N.: Wide residual networks. In: BMVC (2016) Zagoruyko, S., Komodakis, N.: Wide residual networks. In: BMVC (2016)
15.
Zurück zum Zitat Huang, G., Liu, Z., van der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: CVPR (2017) Huang, G., Liu, Z., van der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: CVPR (2017)
17.
Zurück zum Zitat Rebuffi, S.A., Bilen, H., Vedaldi, A.: Learning multiple visual domains with residual adapters. In: NIPS (2017) Rebuffi, S.A., Bilen, H., Vedaldi, A.: Learning multiple visual domains with residual adapters. In: NIPS (2017)
18.
Zurück zum Zitat Bilen, H., Vedaldi, A.: Integrated perception with recurrent multi-task neural networks. In: NIPS (2016) Bilen, H., Vedaldi, A.: Integrated perception with recurrent multi-task neural networks. In: NIPS (2016)
20.
Zurück zum Zitat Kokkinos, I.: Ubernet: training a universal convolutional neural network for low-, mid-, and high-level vision using diverse datasets and limited memory. In: CVPR (2017) Kokkinos, I.: Ubernet: training a universal convolutional neural network for low-, mid-, and high-level vision using diverse datasets and limited memory. In: CVPR (2017)
21.
Zurück zum Zitat Shmelkov, K., Schmid, C., Alahari, K.: Incremental learning of object detectors without catastrophic forgetting. In: ICCV (2017) Shmelkov, K., Schmid, C., Alahari, K.: Incremental learning of object detectors without catastrophic forgetting. In: ICCV (2017)
22.
Zurück zum Zitat Lee, S.W., Kim, J.H., Ha, J.W., Zhang, B.T.: Overcoming catastrophic forgetting by incremental moment matching. In: NIPS (2017) Lee, S.W., Kim, J.H., Ha, J.W., Zhang, B.T.: Overcoming catastrophic forgetting by incremental moment matching. In: NIPS (2017)
23.
Zurück zum Zitat Han, S., Pool, J., Tran, J., Dally, W.: Learning both weights and connections for efficient neural network. In: NIPS (2015) Han, S., Pool, J., Tran, J., Dally, W.: Learning both weights and connections for efficient neural network. In: NIPS (2015)
26.
Zurück zum Zitat Courbariaux, M., Bengio, Y., David, J.P.: BinaryConnect: training deep neural networks with binary weights during propagations. In: NIPS (2015) Courbariaux, M., Bengio, Y., David, J.P.: BinaryConnect: training deep neural networks with binary weights during propagations. In: NIPS (2015)
27.
Zurück zum Zitat Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., Bengio, Y.: Binarized neural networks. In: NIPS (2016) Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., Bengio, Y.: Binarized neural networks. In: NIPS (2016)
29.
Zurück zum Zitat Zhu, C., Han, S., Mao, H., Dally, W.J.: Trained ternary quantization. In: ICLR (2017) Zhu, C., Han, S., Mao, H., Dally, W.J.: Trained ternary quantization. In: ICLR (2017)
30.
Zurück zum Zitat Guo, Y., Yao, A., Chen, Y.: Dynamic network surgery for efficient DNNs. In: NIPS (2016) Guo, Y., Yao, A., Chen, Y.: Dynamic network surgery for efficient DNNs. In: NIPS (2016)
31.
Zurück zum Zitat Zhou, B., Lapedriza, A., Khosla, A., Oliva, A., Torralba, A.: Places: A 10 million image database for scene recognition. TPAMI (2017) Zhou, B., Lapedriza, A., Khosla, A., Oliva, A., Torralba, A.: Places: A 10 million image database for scene recognition. TPAMI (2017)
32.
Zurück zum Zitat Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: CVPR (2015) Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: CVPR (2015)
Metadaten
Titel
Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weights
verfasst von
Arun Mallya
Dillon Davis
Svetlana Lazebnik
Copyright-Jahr
2018
DOI
https://doi.org/10.1007/978-3-030-01225-0_5