mlsanityData to 7 Oct 2026

Boosting Binary Masks for Multi-Domain Learning through Affine Transformations

arXiv 2103.13894 ↗First posted 25 Mar 2021cs.CVcited by 5

Not on the map. Why?

Abstract (arXiv)

In this work, we present a new, algorithm for multi-domain learning. Given a pretrained architecture and a set of visual domains received sequentially, the goal of multi-domain learning is to produce a single model performing a task in all the domains together. Recent works showed how we can address this problem by masking the internal weights of a given original conv-net through learned binary variables. In this work, we provide a general formulation of binary mask based models for multi-domain learning by affine transformations of the original network parameters. Our formulation obtains significantly higher levels of adaptation to new domains, achieving performances comparable to domain-specific models while requiring slightly more than 1 bit per network parameter per additional domain. Experiments on two popular benchmarks showcase the power of our approach, achieving performances close to state-of-the-art methods on the Visual Decathlon Challenge.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 46 of its 71 references that are in the map. First posted on arXiv 25 Mar 2021.

Builds on

46 of 71 references.

References · 46

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  6. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  7. 2016Densely Connected Convolutional Networkscited by 6,691
  8. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  9. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  10. 2016Wide Residual Networkscited by 3,057
  11. 2016Learning without Forgettingcited by 2,465
  12. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,313
34 more
  1. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  2. 2013Describing Textures in the Wildcited by 2,110
  3. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
  4. 2013Fine-Grained Visual Classification of Aircraftcited by 1,701
  5. 2016Progressive Neural Networkscited by 1,332
  6. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 882
  7. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 829
  8. 2013An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networkscited by 711
  9. 2018Taskonomy: Disentangling Task Transfer Learningcited by 631
  10. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
  11. 2017Learning multiple visual domains with residual adapterscited by 553
  12. 2015Towards Open Set Deep Networkscited by 533
  13. 2018End-to-End Multi-Task Learning with Attentioncited by 533
  14. 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 413
  15. 2014Towards Open World Recognitioncited by 226
  16. 2020Modeling the Background for Incremental Learning in Semantic Segmentationcited by 189
  17. 2015Large-scale Classification of Fine-Art Paintings: Learning The Right Metric on The Right Featurecited by 181
  18. 2018Efficient parametrization of multi-domain deep neural networkscited by 168
  19. 2018SpotTune: Transfer Learning through Adaptive Fine-tuningcited by 166
  20. 2015Adding Gradient Noise Improves Learning for Very Deep Networkscited by 144
  21. 2015Fixed Point Quantization of Deep Convolutional Networkscited by 141
  22. 2017Incremental Learning Through Deep Adaptationcited by 108
  23. 2016Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1cited by 102
  24. 2015Neural Networks with Few Multiplicationscited by 46
  25. 2017Universal representations:The missing link between faces, text, planktons, and cat breedscited by 45
  26. 2018Adding New Tasks to a Single Network with Weight Transformations using Binary Maskscited by 21
  27. 2019Budget-Aware Adapters for Multi-Domain Learningcited by 18
  28. 2019Efficient Multi-Domain Network Learning by Covariance Normalizationcited by 16
  29. 2019Knowledge is Never Enough: Towards Web Aided Deep Open World Recognitioncited by 12
  30. 2018Piggyback: Adapting a Single Network to Multiple Tasks by Learning to Mask Weightscited by 11
  31. 2019NetTailor: Tuning the Architecture, Not Just the Weightscited by 11
  32. 2020Boosting Deep Open World Recognition by Clusteringcited by 9
  33. 2014Scalable Greedy Algorithms for Transfer Learningcited by 2
  34. 2019The RGB-D Triathlon: Towards Agile Visual Toolboxes for Robotscited by 1

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 3.

5 papers.

  1. 2021Cluster-driven Graph Federated Learning over Multiple Domainscited by 43
  2. 2023On the Effectiveness of LayerNorm Tuning for Continual Learning in Vision Transformerscited by 14
  3. 2022Budget-Aware Pruning for Multi-Domain Learningcited by 1
  4. 2023Decoupled Training: Return of Frustratingly Easy Multi-Domain Learningcited by 1
  5. 2023Budget-Aware Pruning: Handling Multiple Domains with Less Parameterscited by 0