Artificial Neural Networks and Machine Learning - ICANN 2021

Adversarial machine learning.- An Improved (Adversarial) Reprogramming Technique for Neural Networks.- Adversarial Robustness in Deep Learning: Attacks on Fragile Neurons.- How to compare adversarial robustness of classifiers from a global perspective.- Multiple-Model based Defense for Deep Reinforcement Learning against Adversarial Attack.- Neural Paraphrase Generation with Multi-Domain Corpus.- Leveraging Adversarial Training to Facilitate Grammatical Error Correction.- Statistical Certi?cation of Acceptable Robustness for Neural Networks.- Model Extraction and Adversarial Attacks on Neural Networks using Switching Power Information.- Anomaly detection.- o 0097 - CmaGraph: A TriBlocks Anomaly Detection Method in Dynamic Graph Using Evolutionary Community Representation Learning.- Falcon: Malware Detection and Categorization with Network Traffic Images.- Attention-based Bi-LSTM for Anomaly Detection on Time-Series Data.- Semi-supervised Graph Edge Convolutional Network for Anomaly Detection.- Feature Creation Towards the Detection of Non-Control-Flow Hijacking Attacks.- Attention and transformers I. - An Attention Module for Convolutional Neural Networks.- Attention-based 3D neural architectures for predicting cracks in designs.- Entity-aware Biaffine Attention for Constituent Parsing.- Attention-based Multi-View Feature Fusion for Cross-Domain Recommendation.- Say in Human-like Way: Hierarchical Cross-modal Information Abstraction and Summarization for Controllable Captioning.- DAEMA: Denoising Autoencoder with Mask Attention.- Spatial-Temporal Traffic Data Imputation via Graph Attention Convolutional Network.- EGAT: Edge-Featured Graph Attention Network.- Attention and transformers II.- Knowledge Graph Enhanced Transformer for Generative Question Answering Tasks.- GAttANet: Global attention agreement for convolutional neural networks.- Classification Models for Partially Ordered Sequences.- TINet: Multi-dimensional Traffic Data Imputation via Transformer Network.- Sequential Self-Attentive model for Knowledge Tracing.- Multi-Object Tracking based on Nearest Optimal Template Library.- TSTNet: A Sequence to Sequence Transformer Network for Spatial-temporal Traffic Prediction.- Audio and multimodal applications.- A multimode two-stream network for egocentric action recognition.- Behavior of Keyword Spotting Networks Under Noisy Conditions.- Robust Stroke Recognition via Vision and IMU in Robotic Table Tennis.- AMVAE: Asymmetric Multimodal Variational Autoencoder for Multi-view Representation.- Enhancing Separate Encoding with Multi-layer Feature Alignment for Image-Text Matching.- Bird Audio Diarization with Faster R-CNN.- Multi-Modal Chorus Recognition for Improving Song Search.- FaVoA: Face-Voice Association Favours Ambiguous Speaker Detection.- Bioinformatics and biosignal analysis.- Identification of Incorrect Karyotypes Using Deep Learning.- A Metagraph-Based Model for Predicting Drug-Target Interaction on Heterogeneous Network.- Evaluating Multiple-Concept Biomedical Hypotheses Based on Deep Sets.- A Network Embedding Based Approach to Drug-Target Interaction Prediction Using Additional Implicit Networks.- Capsule networks.- CNNapsule: A Lightweight Network with Fusion Features for Monocular Depth Estimation.- Learning Optimal Primary Capsules by Information Bottleneck.- Capsule Networks with Routing Annealing.- Training Deep Capsule Networks with Residual Connections.- Cognitive models.- Interpretable Visual Understanding with Cognitive Attention Network.- A Bio-Inspired Mechanism Based on Neural Threshold Regulation to Compensate Variability in Network Connectivity.- A Predictive Coding Account for Chaotic Itinerancy.- A Computational Model of the Effect of Short-Term Monocular Deprivation on Binocular Rivalry in the Context of Amblyopia.- Transitions among metastable states underlie context-dependent working memories in a multiple timescale network.