Information Dropout: learning optimal representations through noisy computation, vol.32, p.23, 2016. ,

What regularized auto-encoders learn from the data-generating distribution, Journal of Machine Learning Research, p.25, 2014. ,

Improving Inception and Image Classification in TensorFlow, p.17, 2016. ,

Deep Variational Information Bottleneck, International Conference on Learning Representations (ICLR, vol.32, p.23, 2017. ,

The effects of adding noise during backpropagation training on a generalization performance, Neural computation (cit, p.20, 1996. ,

Mggan: Solving mode collapse using manifold guided training, p.82, 2018. ,

A theory of learning from different domains, Machine learning, p.106, 2010. ,

Representation learning: A review and new perspectives, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) (cit, p.15, 2013. ,

Greedy layer-wise training of deep networks, Advances in Neural Information Processing Systems (NIPS) (cit, vol.45, p.23, 2007. ,

MU-TAN: Multimodal Tucker Fusion for Visual Question Answering, IEEE International Conference on Computer Vision (ICCV), 2017. ,

The perception-distortion tradeoff, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.81, 2018. ,

Study on training methods and generalization performance of deep learning for image classification, vol.13, p.11, 2018. ,

SHADE: Information Based Regularization for Deep Learning -Extended version, p.31, 2018. ,

SHADE: Information-Based Regularization for Deep Learning, IEEE International Conference on Image Processing (ICIP) (cit. on, vol.11, p.9, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-01994740

Combining labeled and unlabeled data with co-training, Annual Conference on Computational Learning Theory (cit, p.24, 1998. ,

Semi-supervised FusedGAN for Conditional Image Generation, European Conference on Computer Vision (ECCV) (cit. on pp. 26, vol.82, p.49, 2018. ,

Large-scale machine learning with stochastic gradient descent, p.15, 2010. ,

Domain separation networks, Advances in Neural Information Processing Systems (NIPS) (cit, p.107, 2016. ,

Large scale gan training for high fidelity natural image synthesis, International Conference on Learning Representations (ICLR) (cit, p.82, 2019. ,

Invariant Scattering Convolution Networks, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) (cit, p.21, 2013. ,

Cross-Modal Retrieval in the Cooking Context: Learning Semantic Text-Image Embeddings, Special Interest Group on Information Retrieval (SIGIR) (cit, p.44, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-01931470

All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp, vol.107, p.77, 2019. ,

Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2017. ,

Isolating sources of disentanglement in variational autoencoders, Advances in Neural Information Processing Systems (NeurIPS), 2018. ,

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets, Advances in Neural Information Processing Systems (NIPS) (cit, p.129, 2016. ,

Discovering hidden factors of variation in deep networks, International Conference on Machine Learning Workshop (ICML-W) (cit. on p, p.27, 2015. ,

Stargan: Unified generative adversarial networks for multi-domain image-to-image translation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.81, 2018. ,

Multi-column deep neural networks for image classification, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.20, 2012. ,

An analysis of single-layer networks in unsupervised feature learning, International Conference on Artificial Intelligence and Statistics (AISTATS) (cit, p.60, 2011. ,

Group equivariant convolutional networks, International Conference on Machine Learning (ICML), 2016. ,

Elements of information theory, 1991. ,

R-FCN: Object detection via region-based fully convolutional networks, Advances in Neural Information Processing Systems (NIPS) (cit, p.44, 2016. ,

Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks, International Conference on Learning Representations (ICLR, 2017. ,

Dataset augmentation in feature space, International Conference on Learning Representations Workshop (ICLR-W) (cit, p.20, 2017. ,

Rotation-invariant convolutional neural networks for galaxy morphology prediction, Monthly notices of the royal astronomical society, 2015. ,

Adversarial Feature Learning, International Conference on Learning Representations (ICLR) (cit, p.82, 2017. ,

Generating images with perceptual similarity metrics based on deep networks, Advances in Neural Information Processing Systems (NIPS) (cit, vol.87, 2016. ,

Adaptive subgradient methods for online learning and stochastic optimization, Journal of Machine Learning Research, p.15, 2011. ,

Adversarially Learned Inference, International Conference on Learning Representations (ICLR) (cit, p.82, 2017. ,

A guide to convolution arithmetic for deep learning, 2016. ,

Learning disentangled joint continuous and discrete representations, Advances in Neural Information Processing Systems (NeurIPS) (cit, p.83, 2018. ,

Deep Architectures in LaTeX, p.16, 2017. ,

Learning a Deep ConvNet for Multi-label Classification with Partial Labels, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.24, 2019. ,

WELDON: Weakly Supervised Learning of Deep Convolutional Neural Networks, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp, vol.38, p.37, 2016. ,

URL : https://hal.archives-ouvertes.fr/hal-01343785

The Art of Data Augmentation, Journal of Computational and Graphical Statistics, 2001. ,

Latent Constraints: Learning to Generate Conditionally from Unconditional Generative Models, International Conference on Learning Representations (ICLR, vol.83, p.82, 2018. ,

Finding beans in burgers: Deep semantic-visual embedding with localization, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.44, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-02171857

Why does unsupervised pre-training help deep learning?, In: Journal of Machine Learning Research, p.23, 2010. ,

Neocognitron: A self-organizing neural network for a mechanism of pattern recognition unaffected by shift in position, Biological Cybernetics, p.4, 1980. ,

Unsupervised Domain Adaptation by Backpropagation, International Conference on Machine Learning (ICML), p.38, 2015. ,

DomainAdversarial Training of Neural Networks, Journal of Machine Learning Research, p.60, 2016. ,

URL : https://hal.archives-ouvertes.fr/hal-01624607

Shake-Shake regularization of 3-branch residual networks, International Conference on Learning Representations Workshop (ICLR-W), 2017. ,

From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) (cit, p.93, 2001. ,

DropBlock: A regularization method for convolutional networks, Advances in Neural Information Processing Systems (NeurIPS) (cit, p.21, 2018. ,

Deep sparse rectifier neural networks, International Conference on Artificial Intelligence and Statistics (AISTATS) (cit, p.25, 2011. ,

URL : https://hal.archives-ouvertes.fr/hal-00752497

Topdown regularization of deep belief networks, Advances in Neural Information Processing Systems (NIPS) (cit, p.25, 2013. ,

URL : https://hal.archives-ouvertes.fr/hal-00947569

The Reversible Residual Network: Backpropagation Without Storing Activations, Advances in Neural Information Processing Systems (NIPS) (cit. on pp. 50, vol.108, p.51, 2017. ,

Deep Learning, vol.21, p.19, 2016. ,

Generative Adversarial Nets, Advances in Neural Information Processing Systems (NIPS) (cit. on pp. 9, vol.25, p.81, 2014. ,

Explaining and harnessing adversarial examples, International Conference on Learning Representations (ICLR) (cit, vol.26, p.20, 2015. ,

Semi-supervised learning by entropy minimization, Advances in neural information processing systems, p.24, 2005. ,

Noise injection: Theoretical prospects, Neural Computation (cit, p.20, 1997. ,

A Two-Step Disentanglement Method, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp. 7, 84, vol.86, p.91, 2018. ,

More Than 500 Hours Of Content Are Now Being Uploaded To YouTube Every Minute, 2019. ,

Deep Residual Learning for Image Recognition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp. 4, 12, vol.17, p.37, 2016. ,

Attgan: Facial attribute editing by only changing what you want, IEEE Transactions on Image Processing (TIP) (cit, p.82, 2019. ,

Towards a Definition of Disentangled Representations, 2018. ,

beta-vae: Learning basic visual concepts with a constrained variational framework, International Conference on Learning Representations (ICLR, 2017. ,

Reducing the dimensionality of data with neural networks". In: Science (cit, vol.46, p.23, 2006. ,

Neural networks for machine learning lecture 6a overview of mini-batch gradient descent (cit, p.15, 2012. ,

Disentangling Factors of Variation by Mixing Them, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.84, 2018. ,

Receptive fields, binocular interaction and functional architecture in the cat's visual cortex, The Journal of physiology, p.16, 1962. ,

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, Journal of Machine Learning Research, 2016. ,

Label Propagation for Deep Semi-supervised Learning, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. ,

, International Conference on Learning Representations (ICLR) (cit. on pp. 50, vol.108, p.51, 2018.

Unsupervised Adversarial Invariance, Advances in Neural Information Processing Systems (NeurIPS) (cit. on pp, vol.27, p.91, 2018. ,

Shakeout: A new regularized deep neural network training scheme, Conference on Artificial Intelligence (AAAI) (cit, p.20, 2016. ,

Progressive growing of gans for improved quality, stability, and variation, International Conference on Learning Representations (ICLR) (cit, p.81, 2017. ,

A style-based generator architecture for generative adversarial networks, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp. 77, vol.81, 2019. ,

Deep learning without poor local minima, Advances in Neural Information Processing Systems (NIPS) (cit. on p. 5), 2016. ,

, Disentangling by factorising". In: arXiv preprint libary, p.83, 2018.

Adam: A method for stochastic optimization, International Conference on Learning Representations (ICLR) (cit, p.15, 2015. ,

Glow: Generative flow with invertible 1x1 convolutions, Advances in Neural Information Processing Systems (NeurIPS) (cit, p.109, 2018. ,

Semi-supervised learning with deep generative models, Advances in Neural Information Processing Systems (NIPS), 2014. ,

Auto-Encoding Variational Bayes, International Conference on Learning Representations (ICLR) (cit. on pp. 25, vol.80, 2013. ,

Learning Latent Subspaces in Variational Autoencoders, Advances in Neural Information Processing Systems (NeurIPS) (cit. on pp. 7, 84, vol.86, p.91, 2018. ,

On convergence and stability of gans, p.81, 2017. ,

Ubernet: Training a universal convolutional neural network for low-, mid-, and high-level vision using diverse datasets and limited memory, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. ,

Learning multiple layers of features from tiny images, vol.60, p.36, 2009. ,

Imagenet classification with deep convolutional neural networks, Advances in Neural Information Processing Systems (NIPS) (cit. on pp, vol.4, pp.15-17, 2012. ,

A Simple Weight Decay Can Improve Generalization, Advances in Neural Information Processing Systems (NIPS), 1992. ,

Regularization for deep learning: A taxonomy, 2017. ,

Deep Convolutional Inverse Graphics Network, Advances in Neural Information Processing Systems (NIPS), p.84, 2015. ,

Temporal Ensembling for Semi-Supervised Learning, International Conference on Learning Representations (ICLR, 2017. ,

Fader Networks:Manipulating Images by Sliding Attributes, Advances in Neural Information Processing Systems (NIPS) (cit. on pp. 7, 27, vol.84, p.91, 2017. ,

URL : https://hal.archives-ouvertes.fr/hal-02275215

Classification using discriminative restricted Boltzmann machines, International Conference on Machine Learning (ICML), 2008. ,

Supervised autoencoders: Improving generalization performance with unsupervised regularizers, Advances in Neural Information Processing Systems (NeurIPS) (cit, vol.87, p.46, 2018. ,

Backpropagation applied to handwritten zip code recognition, Neural computation (cit, p.4, 1989. ,

Gradient based learning applied to document recognition, Proceedings of the IEEE, vol.60, p.16, 1998. ,

Learning methods for generic object recognition with invariance to pose and lighting, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), p.93, 2004. ,

Photo-realistic single image super-resolution using a generative adversarial network, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.81, 2017. ,

Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks, International Conference on Machine Learning Workshop (ICML-W) (cit, p.24, 2013. ,

, Layer normalization". In: arXiv preprint libary, p.22, 2016.

, Whiteout: Gaussian Adaptive Noise Regularization in FeedForward Neural Networks". In: arXiv preprint libary, p.20, 2016.

Multi-Task Adversarial Network for Disentangled Feature Learning, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp, vol.27, p.91, 2018. ,

Exploring Disentangled Feature Representation Beyond Face Identification, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit. on pp. 7, 27, vol.84, 2018. ,

Deep Learning Face Attributes in the Wild, IEEE International Conference on Computer Vision (ICCV) (cit, p.92, 2015. ,

The variational fair autoencoder, International Conference on Learning Representations (ICLR) (cit, p.77, 2016. ,

The expressive power of neural networks: A view from the width, Advances in Neural Information Processing Systems (NIPS) (cit. on p, vol.18, 2017. ,

Adversarial training of partially invertible variational autoencoders, 2019. ,

URL : https://hal.archives-ouvertes.fr/hal-01886285

Understanding the effective receptive field in deep convolutional neural networks, Advances in Neural Information Processing Systems (NIPS) (cit, p.16, 2016. ,

Human pose regression by combining indirect part detection and contextual information, p.54, 2017. ,

Adversarial Autoencoders, International Conference on Learning Representations (ICLR) (cit, p.82, 2016. ,

Group Invariant Scattering, Communications on Pure and Applied Mathematics (CPAM) (cit. on p. 21), 2012. ,

A wavelet tour of signal processing : the sparse way, p.52, 2009. ,

Disentangling factors of variation in deep representation using adversarial training, Advances in Neural Information Processing Systems (NIPS) (cit. on pp. 7, 27, vol.77, 2016. ,

A logical calculus of the ideas immanent in nervous activity, The bulletin of mathematical biophysics, p.15, 1943. ,

Manifold Learning in Quotient Spaces, 2018. ,

Distributional smoothing with virtual adversarial training, International Conference on Learning Representations (ICLR) (cit, p.26, 2016. ,

Adding gradient noise improves learning for very deep networks, International Conference on Learning Representations (ICLR, p.20, 2016. ,

Reading digits in natural images with unsupervised feature learning, Advances in Neural Information Processing Systems Workshop (NIPS-W), p.60, 2011. ,

Regularizing deep neural networks by noise: Its interpretation and optimization, Advances in Neural Information Processing Systems (NIPS) (cit, p.20, 2017. ,

The Top 20 Valuable Facebook Statistics, 2019. ,

Activation functions: Comparison of trends in practice and research for deep learning, p.15, 2018. ,

Feature Visualization, 2017. ,

Realistic evaluation of deep semi-supervised learning algorithms, Advances in Neural Information Processing Systems (NeurIPS) (cit, p.97, 2018. ,

Estimation of Entropy and Mutual Information, Neural Computation, p.129, 2003. ,

Jigsaw Puzzle Solving Using Local Feature Co-Occurrences in Deep Neural Networks, IEEE International Conference on Image Processing (ICIP) (cit, p.53, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-01820489

Reconstruction-based disentanglement for pose-invariant face recognition, IEEE International Conference on Computer Vision (ICCV) (cit. on pp, vol.84, p.27, 2017. ,

Invertible conditional gans for image editing, Advances in Neural Information Processing Systems Workshop (NIPS-W) (cit. on pp. 7, 27, vol.77, p.91, 2016. ,

Regularizing Neural Networks by Penalizing Confident Output Distributions, International Conference on Learning Representations Workshop (ICLR-W), p.23, 2017. ,

Experiments on Learning by Back Propagation, p.20, 1986. ,

Sparse Feature Learning for Deep Belief Networks, Advances in Neural Information Processing Systems (NIPS) (cit, p.25, 2008. ,

Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.49, 2007. ,

Efficient Learning of Sparse Representations with an Energy-Based Model, Advances in Neural Information Processing Systems (NIPS) (cit, p.25, 2007. ,

Semi-supervised learning of compact document representations with deep networks, International Conference on Machine Learning (ICML), 2008. ,

Semi-supervised learning with ladder networks, Advances in Neural Information Processing Systems (NIPS) (cit. on pp, vol.46, p.71, 2015. ,

Faster R-CNN: Towards real-time object detection with region proposal networks, Advances in Neural Information Processing Systems (NIPS), 2015. ,

Contractive auto-encoders: Explicit invariance during feature extraction, International Conference on Machine Learning (ICML), p.22, 2011. ,

HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning, European Conference on Computer Vision (ECCV) (cit. on pp. 9, vol.157, p.43, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-02073640

DualDis: DualBranch Disentangling with Adversarial Learning, Under Review at Advances in Neural Information Processing Systems (NeurIPS) (cit. on, vol.75, p.9, 2019. ,

Variational approaches for auto-encoding generative adversarial networks, p.82, 2017. ,

Stabilizing training of generative adversarial networks through regularization, Advances in Neural Information Processing Systems (NIPS) (cit, p.81, 2017. ,

Learning Disentangled Representations with Reference-Based Variational Autoencoders, p.77, 2019. ,

URL : https://hal.archives-ouvertes.fr/hal-01896007

Learning representations by back-propagating errors, Cognitive modeling, 1988. ,

ImageNet Large Scale Visual Recognition Challenge, International Journal of Computer Vision, vol.37, 2015. ,

Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning, Advances in Neural Information Processing Systems (NIPS), 2016. ,

Improved Techniques for Training GANs, Advances in Neural Information Processing Systems (NIPS), 2016. ,

How does batch normalization help optimization, Advances in Neural Information Processing Systems (NeurIPS), 2018. ,

L1-norm double backpropagation adversarial defense, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), p.22, 2019. ,

URL : https://hal.archives-ouvertes.fr/hal-02049020

Learning and generalization with the information bottleneck, Theoretical Computer Science (cit, p.23, 2010. ,

Transductive semi-supervised deep learning using min-max features, Proceedings of the European Conference on Computer Vision (ECCV), p.24, 2018. ,

Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance, Riza Alp Guler, Dimitris Samaras, Nikos Paragios, and Iasonas Kokkinos, p.54, 2018. ,

URL : https://hal.archives-ouvertes.fr/hal-01935596

Neural face editing with intrinsic image disentangling, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.54, 2017. ,

Best practices for convolutional neural networks applied to visual document analysis, Icdar (cit, p.20, 2003. ,

Very deep convolutional networks for large-scale image recognition, International Conference on Learning Representations (ICLR) (cit. on pp, vol.23, pp.15-18, 2015. ,

Robust large margin deep neural networks, IEEE Transactions on Signal Processing, p.22, 2017. ,

Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks, International Conference on Learning Representations (ICLR, 2016. ,

Dropout: A Simple Way to Prevent Neural Networks from Overfitting, Journal of Machine Learning Research, 2014. ,

, The Lifting Scheme: A New Philosophy in Biorthogonal Wavelet Constructions". In: Wavelet Applications in Signal and Image Processing III, p.50, 1995.

Going deeper with convolutions, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, vol.18, p.17, 2015. ,

Rethinking the inception architecture for computer vision, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, vol.20, p.17, 2016. ,

Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results, Advances in Neural Information Processing Systems (NIPS), vol.66, pp.140-142, 2017. ,

The information bottleneck method, Annual Allerton Conference on Communication, Control and Computing, 1999. ,

Deep learning and the information bottleneck principle, Information Theory Workshop (ITW). IEEE, vol.128, p.31, 2015. ,

Disentangled Representation Learning GAN for Pose-Invariant Face Recognition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.84, 2017. ,

Word representations: a simple and general method for semi-supervised learning, Proceedings of the 48th annual meeting of the association for computational linguistics, 2010. ,

Instance normalization: The missing ingredient for fast stylization, p.22, 2016. ,

Principles of risk minimization for learning theory, Advances in Neural Information Processing Systems (NIPS) (cit. on p, vol.18, 1992. ,

On the uniform convergence of relative frequencies of events to their probabilities, Measures of Complexity, p.18, 1972. ,

Extracting and composing robust features with denoising autoencoders, International Conference on Machine Learning (ICML), p.25, 2008. ,

Regularization of Neural Networks using DropConnect, International Conference on Machine Learning (ICML), 2013. ,

High-resolution image synthesis and semantic manipulation with conditional gans, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (cit, p.82, 2018. ,

Deep Learning via Semi-supervised Embedding, International Conference on Machine Learning (ICML), 2008. ,

The Devil is in the Decoder, British Machine Vision Conference (BMVC) (cit, p.50, 2017. ,

Group normalization, European Conference on Computer Vision (ECCV) (cit, p.22, 2018. ,

, Wide Residual Networks". In: arXiv preprint libary, p.36, 2016.

URL : https://hal.archives-ouvertes.fr/hal-01832503

Visualizing and understanding convolutional networks, European Conference on Computer Vision (ECCV) (cit, p.55, 2014. ,

Understanding deep learning requires rethinking generalization, International Conference on Learning Representations (ICLR, 2017. ,

Fixup Initialization: Residual Learning Without Normalization, International Conference on Learning Representations (ICLR) (cit. on p. 5), 2019. ,

Augmenting Supervised Neural Networks with Unsupervised Objectives for Large-scale Image Classification, International Conference on Machine Learning (ICML) (cit. on pp. 23, vol.46, p.55, 2016. ,

Stacked What-Where Auto-encoders, International Conference on Learning Representations Workshop (ICLR-W) (cit. on pp, vol.53, p.66, 2016. ,

Unpaired image-to-image translation using cycle-consistent adversarial networks, IEEE International Conference on Computer Vision (ICCV) (cit, p.81, 2017. ,

Sparsely aggregated convolutional networks, European Conference on Computer Vision (ECCV) (cit, p.22, 2018. ,

Semi-Supervised Learning Literature Survey, p.24, 2005. ,

Learning from labeled and unlabeled data with label propagation, p.24, 2002. ,

, unlabeled images) with batches of 11 unlabeled images and 5 labeled images. Hyperparameters values and scheduling over training are detailed in Table B, vol.8

, ? In the encoder, every layer is followed by batch normalization and ReLU

, ? In the decoder, every layer is followed by a batch normalization and LeakyReLU(0.2), except last layer which has no activation or BN

, ? In the classifiers, every intermediate layer is followed by a ReLU

, ? Layers are described using the following syntax: ? Conv: 128