Artificial Neural Networks and Machine Learning – ICANN 2019: Deep Learning [electronic resource] : 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part II / edited by Igor V. Tetko, Věra Kůrková, Pavel Karpov, Fabian Theis.

Adaptive Graph Fusion for Unsupervised Feature Selection -- Unsupervised Feature Selection via Local Total-order Preservation -- Discrete Stochastic Search and its Application to Feature-Selection for Deep Relational Machines -- Joint Dictionary Learning for Unsupervised Feature Selection -- Comparison between Filter Criteria for Feature Selection in Regression -- CancelOut: A layer for feature selection in deep neural networks -- Adaptive-L2 Batch Neural Gas -- Application of Self Organizing Map to Preprocessing Input Vectors for Convolutional Neural Network -- Hierarchical Reinforcement Learning with Unlimited Recursive Subroutine Calls -- Automatic Augmentation by Hill Climbing -- Learning Camera-invariant Representation for Person Re-identification -- PA-RetinaNet: Path Augmented RetinaNet for Dense Object Detection -- Singular Value Decomposition and Neural Networks -- PCI: Principal Component Initialization for Deep Autoencoders -- Improving Weight Initialization of ReLU and Output Layers -- Post-synaptic potential regularization has potential -- A Novel Modification on the Levenberg-Marquardt Algorithm for Avoiding Overfitting in Neural Network Training -- Sign Based Derivative Filtering for Stochastic Gradient Descent -- Architecture-aware Bayesian Optimization for Neural Network Tuning -- Non-Convergence and Limit Cycles in the Adam Optimizer -- Learning Internal Dense But External Sparse Structures of Deep Convolutional Neural Network -- Using feature entropy to guide filter pruning for efficient convolutional networks -- Simultaneously Learning Architectures and Features of Deep Neural Networks -- Learning Sparse Hidden States in Long Short-Term Memory -- Multi-objective Pruning for CNNs using Genetic Algorithm -- Dynamically Sacrificing Accuracy for Reduced Computation: Cascaded Inference Based on Softmax Confidence -- Light-Weight Edge Enhanced Network for On-orbit Semantic Segmentation -- Local Normalization Based BN Layer Pruning -- On Practical Approach to Uniform Quantization of Non-redundant Neural Networks -- Residual learning for FC kernels of convolutional network -- A Novel Neural Network-based Symbolic Regression Method: Neuro-Encoded Expression Programming -- Compute-efficient neural network architecture optimization by a genetic algorithm -- Controlling Model Complexity in Probabilistic Model-Based Dynamic Optimization of Neural Network Structures -- Predictive Uncertainty Estimation with Temporal Convolutional Networks for Dynamic Evolutionary Optimization -- Sparse Recurrent Mixture Density Networks for Forecasting High Variability Time Series with Confidence Estimates -- A multitask learning neural network for short-term traffic speed prediction and confidence estimation -- Central-diffused Instance Generation Method in Class Incremental Learning -- Marginal Replay vs Conditional Replay for Continual Learning -- Simplified computation and interpretation of Fisher matrices in incremental learning with deep neural networks -- Active Learning for Image Recognition using a Visualization-Based User Interface -- Basic Evaluation Scenarios for Incrementally Trained Classifiers -- Embedding Complexity of Learned Representations in Neural Networks -- Joint Metric Learning on Riemannian Manifold of Global Gaussian Distributions -- Multi-Task Sparse Regression Metric Learning for Heterogeneous Classification -- Fast Approximate Geodesics for Deep Generative Models -- Spatial Attention Network for Few-Shot Learning -- Routine Modeling with Time Series Metric Learning -- Leveraging Domain Knowledge for Reinforcement Learning using MMC Architectures -- Conditions for Unnecessary Logical Constraints in Kernel Machines -- HiSeqGAN: Hierarchical Sequence Synthesis and Prediction -- DeepEX: Bridging the Gap Between Knowledge and Data Driven Techniques for Time Series Forecasting -- Transferable Adversarial Cycle Alignment for Domain Adaption -- Evaluation of domain adaptation approaches for robust classification of heterogeneous biological data sets -- Named Entity Recognition for Chinese Social Media with Domain Adversarial Training and Language Modeling -- Deep Domain Knowledge Distillation for Person Re-identification -- A study on catastrophic forgetting in deep LSTM networks -- A Label-specific Attention-based Network with Regularized Loss for Multi-label Classification -- An Empirical Study of Multi-domain and Multi-task Learning in Chinese Named Entity Recognition -- Filter Method Ensemble with Neural Networks -- Dynamic Centroid Insertion and Adjustment for Data Sets with Multiple Imbalanced Classes -- Increasing the Generalisaton Capacity of Conditional VAEs -- Playing the Large Margin Preference Game.

The proceedings set LNCS 11727, 11728, 11729, 11730, and 11731 constitute the proceedings of the 28th International Conference on Artificial Neural Networks, ICANN 2019, held in Munich, Germany, in September 2019. The total of 277 full papers and 43 short papers presented in these proceedings was carefully reviewed and selected from 494 submissions. They were organized in 5 volumes focusing on theoretical neural computation; deep learning; image processing; text and time series; and workshop and special sessions. .