14.5.7.5.3 Forgetting, Explaination, Intrepretation, Understanding of Convolutional Neural Networks

Chapter Contents (Back)
Convolutional Neural Networks. Forgetting.

Chung, F.L., Wang, S., Deng, Z., Hu, D.,
CATSMLP: Toward a Robust and Interpretable Multilayer Perceptron With Sigmoid Activation Functions,
SMC-B(36), No. 6, December 2006, pp. 1319-1331.
IEEE DOI 0701
BibRef

Peharz, R.[Robert], Gens, R.[Robert], Pernkopf, F.[Franz], Domingos, P.[Pedro],
On the Latent Variable Interpretation in Sum-Product Networks,
PAMI(39), No. 10, October 2017, pp. 2030-2044.
IEEE DOI 1709
Bayes methods, Computational modeling, Inference algorithms, Mixture models, Periodic structures, Probabilistic logic, Semantics, MPE inference, Sum-product networks, expectation-maximization, latent variables, mixture, models See also Sum-product networks: A new deep architecture. BibRef

Montavon, G.[Grégoire], Lapuschkin, S.[Sebastian], Binder, A.[Alexander], Samek, W.[Wojciech], Müller, K.R.[Klaus-Robert],
Explaining nonlinear classification decisions with deep Taylor decomposition,
PR(65), No. 1, 2017, pp. 211-222.
Elsevier DOI 1702
Deep neural networks BibRef

Lapuschkin, S., Binder, A., Montavon, G.[Grégoire], Müller, K.R.[Klaus-Robert], Samek, W.[Wojciech],
Analyzing Classifiers: Fisher Vectors and Deep Neural Networks,
CVPR16(2912-2920)
IEEE DOI 1612
BibRef

Samangouei, P.[Pouya], Saeedi, A.[Ardavan], Nakagawa, L.[Liam], Silberman, N.[Nathan],
ExplainGAN: Model Explanation via Decision Boundary Crossing Transformations,
ECCV18(X: 681-696).
Springer DOI 1810
BibRef

Li, Z.Z.[Zhi-Zhong], Hoiem, D.[Derek],
Learning Without Forgetting,
PAMI(40), No. 12, December 2018, pp. 2935-2947.
IEEE DOI 1811
BibRef
Earlier: ECCV16(IV: 614-629).
Springer DOI 1611
Keep the old results in NN, but learn new capability. Feature extraction, Deep learning, Training data, Neural networks, Convolutional neural networks, Knowledge engineering, visual recognition. BibRef

Mopuri, K.R., Garg, U., Babu, R.V.[R. Venkatesh],
CNN Fixations: An Unraveling Approach to Visualize the Discriminative Image Regions,
IP(28), No. 5, May 2019, pp. 2116-2125.
IEEE DOI 1903
computer vision, convolutional neural nets, feature extraction, object recognition, CNN fixations, discriminative image regions, weakly supervised localization BibRef

Kuo, C.C.J.[C.C. Jay], Zhang, M.[Min], Li, S.[Siyang], Duan, J.[Jiali], Chen, Y.[Yueru],
Interpretable convolutional neural networks via feedforward design,
JVCIR(60), 2019, pp. 346-359.
Elsevier DOI 1903
Interpretable machine learning, Convolutional neural networks, Principal component analysis, Dimension reduction BibRef

Li, H.[Heyi], Tian, Y.[Yunke], Mueller, K.[Klaus], Chen, X.[Xin],
Beyond saliency: Understanding convolutional neural networks from saliency prediction on layer-wise relevance propagation,
IVC(83-84), 2019, pp. 70-86.
Elsevier DOI 1904
Convolutional neural networks, Deep learning understanding, Salient relevance map, Attention area BibRef

Fan, C.X.[Chun-Xiao], Li, Y.[Yang], Tian, L.[Lei], Li, Y.[Yong],
Rectifying Transformation Networks for Transformation-Invariant Representations with Power Law,
IEICE(E102-D), No. 3, March 2019, pp. 675-679.
WWW Link. 1904
CNN to rectify learned feature representations. BibRef

Cao, C.S.[Chun-Shui], Huang, Y.Z.[Yong-Zhen], Yang, Y.[Yi], Wang, L.[Liang], Wang, Z.L.[Zi-Lei], Tan, T.N.[Tie-Niu],
Feedback Convolutional Neural Network for Visual Localization and Segmentation,
PAMI(41), No. 7, July 2019, pp. 1627-1640.
IEEE DOI 1906
Neurons, Visualization, Image segmentation, Semantics, Convolutional neural networks, Task analysis, object segmentation BibRef

Zhou, B.[Bolei], Bau, D.[David], Oliva, A.[Aude], Torralba, A.[Antonio],
Interpreting Deep Visual Representations via Network Dissection,
PAMI(41), No. 9, Sep. 2019, pp. 2131-2145.
IEEE DOI 1908
Method quantifies the interpretability of CNN representations. Visualization, Detectors, Training, Image color analysis, Task analysis, Image segmentation, Semantics, interpretable machine learning BibRef

Liu, R.S.[Ri-Sheng], Cheng, S.C.[Shi-Chao], Ma, L.[Long], Fan, X.[Xin], Luo, Z.X.[Zhong-Xuan],
Deep Proximal Unrolling: Algorithmic Framework, Convergence Analysis and Applications,
IP(28), No. 10, October 2019, pp. 5013-5026.
IEEE DOI 1909
Task analysis, Optimization, Convergence, Mathematical model, Network architecture, Computer architecture, Data models, low-level computer vision BibRef

Cui, X.R.[Xin-Rui], Wang, D.[Dan], Wang, Z.J.[Z. Jane],
Multi-Scale Interpretation Model for Convolutional Neural Networks: Building Trust Based on Hierarchical Interpretation,
MultMed(21), No. 9, September 2019, pp. 2263-2276.
IEEE DOI 1909
Visualization, Computational modeling, Analytical models, Feature extraction, Perturbation methods, Image segmentation, model-agnostic BibRef

Wang, W.[Wei], Zhu, L.[Liqiang], Guo, B.[Baoqing],
Reliable identification of redundant kernels for convolutional neural network compression,
JVCIR(63), 2019, pp. 102582.
Elsevier DOI 1909
Network compression, Convolutional neural network, Pruning criterion, Channel-level pruning BibRef

Hu, S.X.[Shell Xu], Zagoruyko, S.[Sergey], Komodakis, N.[Nikos],
Exploring weight symmetry in deep neural networks,
CVIU(187), 2019, pp. 102786.
Elsevier DOI 1909
BibRef


Buhrmester, V.[Vanessa], Münch, D.[David], Bulatov, D.[Dimitri], Arens, M.[Michael],
Evaluating the Impact of Color Information in Deep Neural Networks,
IbPRIA19(I:302-316).
Springer DOI 1910
BibRef

de la Calle, A.[Alejandro], Tovar, J.[Javier], Almazán, E.J.[Emilio J.],
Geometric Interpretation of CNNs' Last Layer,
IbPRIA(I:137-147).
Springer DOI 1910
BibRef

Rio-Torto, I.[Isabel], Fernandes, K.[Kelwin], Teixeira, L.F.[Luís F.],
Towards a Joint Approach to Produce Decisions and Explanations Using CNNs,
IbPRIA(I:3-15).
Springer DOI 1910
BibRef

El Khatib, A.[Alaa], Karray, F.[Fakhri],
Strategies for Improving Single-Head Continual Learning Performance,
ICIAR19(I:452-460).
Springer DOI 1909
Forgetting. Problem is also not all data is available at once. BibRef

Kamma, K.[Koji], Isoda, Y.[Yuki], Inoue, S.[Sarimu], Wada, T.[Toshikazu],
Behavior-Based Compression for Convolutional Neural Networks,
ICIAR19(I:427-439).
Springer DOI 1909
Reducing redundancy. BibRef

Tartaglione, E.[Enzo], Grangetto, M.[Marco],
Take a Ramble into Solution Spaces for Classification Problems in Neural Networks,
CIAP19(I:345-355).
Springer DOI 1909
BibRef

Gu, J.D.[Jin-Dong], Yang, Y.C.[Yin-Chong], Tresp, V.[Volker],
Understanding Individual Decisions of CNNs via Contrastive Backpropagation,
ACCV18(III:119-134).
Springer DOI 1906
BibRef

Yu, T.[Tao], Long, H.[Huan], Hopcroft, J.E.[John E.],
Curvature-based Comparison of Two Neural Networks,
ICPR18(441-447)
IEEE DOI 1812
Manifolds, Biological neural networks, Tensile stress, Measurement, Matrix decomposition, Covariance matrices BibRef

Malakhova, K.[Katerina],
Representation of Categories in Filters of Deep Neural Networks,
Cognitive18(2054-20542)
IEEE DOI 1812
Visualization, Face, Feature extraction, Detectors, Biological neural networks, Neurons, Automobiles BibRef

Wang, Y.S.[Yi-Sen], Liu, W.Y.[Wei-Yang], Ma, X.J.[Xing-Jun], Bailey, J.[James], Zha, H.Y.[Hong-Yuan], Song, L.[Le], Xia, S.T.[Shu-Tao],
Iterative Learning with Open-set Noisy Labels,
CVPR18(8688-8696)
IEEE DOI 1812
Noise measurement, Feature extraction, Cats, Training, Training data, Labeling, Convolutional neural networks BibRef

Hayes, T.L., Kemker, R., Cahill, N.D., Kanan, C.,
New Metrics and Experimental Paradigms for Continual Learning,
DeepLearnRV18(2112-21123)
IEEE DOI 1812
Robots, Measurement, Training, Task analysis, Computational modeling, Neural networks, Data models BibRef

Kanbak, C.[Can], Moosavi-Dezfooli, S.M.[Seyed-Mohsen], Frossard, P.[Pascal],
Geometric Robustness of Deep Networks: Analysis and Improvement,
CVPR18(4441-4449)
IEEE DOI 1812
Robustness, Manifolds, Additives, Training, Atmospheric measurements, Particle measurements BibRef

Shmelkov, K., Schmid, C., Alahari, K.,
Incremental Learning of Object Detectors without Catastrophic Forgetting,
ICCV17(3420-3429)
IEEE DOI 1802
learning (artificial intelligence), neural nets, object detection, COCO datasets, PASCAL VOC 2007, annotations, Training data BibRef

Rannen, A.[Amal], Aljundi, R.[Rahaf], Blaschko, M.B.[Matthew B.], Tuytelaars, T.[Tinne],
Encoder Based Lifelong Learning,
ICCV17(1329-1337)
IEEE DOI 1802
Learning usually adapts to the most recent task, need a sequence of tasks. feature extraction, image classification, learning (artificial intelligence), catastrophic forgetting, Training BibRef

Aljundi, R.[Rahaf], Babiloni, F.[Francesca], Elhoseiny, M.[Mohamed], Rohrbach, M.[Marcus], Tuytelaars, T.[Tinne],
Memory Aware Synapses: Learning What (not) to Forget,
ECCV18(III: 144-161).
Springer DOI 1810
BibRef

Liu, X.L.[Xia-Lei], Masana, M., Herranz, L., van de Weijer, J.[Joost], López, A.M., Bagdanov, A.D.[Andrew D.],
Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgetting,
ICPR18(2262-2268)
IEEE DOI 1812
Task analysis, Training, Training data, Neural networks, Data models, Computer vision, Standards BibRef

Fawzi, A., Moosavi-Dezfooli, S., Frossard, P., Soatto, S.,
Empirical Study of the Topology and Geometry of Deep Networks,
CVPR18(3762-3770)
IEEE DOI 1812
Neural networks, Perturbation methods, Geometry, Network topology, Topology, Robustness, Optimization BibRef

Zhang, Z.M.[Zi-Ming], Wu, Y.W.[Yuan-Wei], Wang, G.H.[Guang-Hui],
BPGrad: Towards Global Optimality in Deep Learning via Branch and Pruning,
CVPR18(3301-3309)
IEEE DOI 1812
Optimization, Linear programming, Upper bound, Approximation algorithms, Biological neural networks, Convergence BibRef

Palacio, S., Folz, J., Hees, J., Raue, F., Borth, D., Dengel, A.,
What do Deep Networks Like to See?,
CVPR18(3108-3117)
IEEE DOI 1812
Image reconstruction, Training, Neural networks, Decoding, Task analysis, Convolution, Image coding BibRef

Aodha, O.M., Su, S., Chen, Y., Perona, P., Yue, Y.,
Teaching Categories to Human Learners with Visual Explanations,
CVPR18(3820-3828)
IEEE DOI 1812
Education, Visualization, Task analysis, Adaptation models, Mathematical model, Computational modeling, Computer vision BibRef

Fong, R., Vedaldi, A.,
Net2Vec: Quantifying and Explaining How Concepts are Encoded by Filters in Deep Neural Networks,
CVPR18(8730-8738)
IEEE DOI 1812
Semantics, Visualization, Image segmentation, Probes, Neural networks, Task analysis, Training BibRef

Mascharka, D., Tran, P., Soklaski, R., Majumdar, A.,
Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning,
CVPR18(4942-4950)
IEEE DOI 1812
Visualization, Cognition, Task analysis, Neural networks, Image color analysis, Knowledge discovery, Automobiles BibRef

Wang, Y., Su, H., Zhang, B., Hu, X.,
Interpret Neural Networks by Identifying Critical Data Routing Paths,
CVPR18(8906-8914)
IEEE DOI 1812
Routing, Logic gates, Neural networks, Predictive models, Encoding, Semantics, Analytical models BibRef

Dong, Y.P.[Yin-Peng], Su, H.[Hang], Zhu, J.[Jun], Zhang, B.[Bo],
Improving Interpretability of Deep Neural Networks with Semantic Information,
CVPR17(975-983)
IEEE DOI 1711
Computational modeling, Decoding, Feature extraction, Neurons, Semantics, Visualization BibRef

Bau, D., Zhou, B., Khosla, A., Oliva, A., Torralba, A.B.,
Network Dissection: Quantifying Interpretability of Deep Visual Representations,
CVPR17(3319-3327)
IEEE DOI 1711
Detectors, Image color analysis, Image segmentation, Semantics, Training, Visualization BibRef

Hu, R.H.[Rong-Hang], Andreas, J.[Jacob], Darrell, T.J.[Trevor J.], Saenko, K.[Kate],
Explainable Neural Computation via Stack Neural Module Networks,
ECCV18(VII: 55-71).
Springer DOI 1810
BibRef

Rupprecht, C., Laina, I., Navab, N., Hager, G.D., Tombari, F.,
Guide Me: Interacting with Deep Networks,
CVPR18(8551-8561)
IEEE DOI 1812
Image segmentation, Visualization, Natural languages, Task analysis, Semantics, Head, Training BibRef

Zhang, Q., Wu, Y.N., Zhu, S.,
Interpretable Convolutional Neural Networks,
CVPR18(8827-8836)
IEEE DOI 1812
Visualization, Semantics, Integrated circuits, Convolutional neural networks, Task analysis, Training, Entropy BibRef

Khan, S.H.[Salman H.], Hayat, M.[Munawar], Porikli, F.M.[Fatih Murat],
Scene Categorization with Spectral Features,
ICCV17(5639-5649)
IEEE DOI 1802
Explain the network results. feature extraction, image classification, image representation, learning (artificial intelligence), natural scenes, transforms, Transforms BibRef

Worrall, D.E.[Daniel E.], Garbin, S.J.[Stephan J.], Turmukhambetov, D.[Daniyar], Brostow, G.J.[Gabriel J.],
Interpretable Transformations with Encoder-Decoder Networks,
ICCV17(5737-5746)
IEEE DOI 1802
I.e. rotation effects. Explain results. decoding, image coding, interpolation, transforms, complex transformation encoding, BibRef

Sankaranarayanan, S.[Swami], Jain, A.[Arpit], Lim, S.N.[Ser Nam],
Guided Perturbations: Self-Corrective Behavior in Convolutional Neural Networks,
ICCV17(3582-3590)
IEEE DOI 1802
Perturb the inputs, understand NN results. Explain. image classification, image representation, neural nets, CIFAR10 datasets, MNIST, PASCAL VOC dataset, Semantics BibRef

Selvaraju, R.R.[Ramprasaath R.], Cogswell, M.[Michael], Das, A.[Abhishek], Vedantam, R.[Ramakrishna], Parikh, D.[Devi], Batra, D.[Dhruv],
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,
ICCV17(618-626)
IEEE DOI 1802
Explain the CNN models. convolution, data visualisation, gradient methods, image classification, image representation, inference mechanisms, Visualization BibRef

Chapter on Pattern Recognition, Clustering, Statistics, Grammars, Learning, Neural Nets, Genetic Algorithms continues in
Convolutional Neural Networks for Object Detection and Segmentation .


Last update:Oct 1, 2019 at 15:23:24