
Machine Learning and Knowledge Discovery in Databases. Research Track
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The multi-volume set LNAI 12975 until 12979 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2021, which was held during September 13-17, 2021. The conference was originally planned to take place in Bilbao, Spain, but changed to an online event due to the COVID-19 pandemic.
The 210 full papers presented in these proceedings were carefully reviewed and selected from a total of 869 submissions.
The volumes are organized in topical sections as follows:
Research Track:
Part I: Online learning; reinforcement learning; time series, streams, and sequence models; transfer and multi-task learning; semi-supervised and few-shot learning; learning algorithms and applications.
Part II: Generative models; algorithms and learning theory; graphs and networks; interpretation, explainability, transparency, safety.
Part III: Generative models; search and optimization; supervised learning; text mining and natural language processing; image processing, computer vision and visual analytics.
Applied Data Science Track:
Part IV: Anomaly detection and malware; spatio-temporal data; e-commerce and finance; healthcare and medical applications (including Covid); mobility and transportation.
Part V: Automating machine learning, optimization, and feature engineering; machine learning based simulations and knowledge discovery; recommender systems and behavior modeling; natural language processing; remote sensing, image and video processing; social media.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents - Part III
- Generative Models
- Deep Conditional Transformation Models
- 1 Introduction
- 1.1 Transformation Models
- 1.2 Related Work and Our Contribution
- 2 Model and Network Definition
- 2.1 Model Definition
- 2.2 Network Definition
- 2.3 Penalization
- 2.4 Bijectivitiy and Monotonocity Constraints
- 2.5 Interpretability and Identifiability Constraints
- 3 Numerical Experiments
- 4 Application
- 4.1 Movie Reviews
- 4.2 UTKFace
- 4.3 Benchmark Study
- 5 Conclusion and Outlook
- References
- Disentanglement and Local Directions of Variance
- 1 Introduction
- 2 Related Work
- 3 Disentanglement, PCA and VAEs
- 3.1 Preliminaries
- 3.2 Disentanglement in a PCA Setting
- 3.3 PCA Behavior in Variational Autoencoders
- 4 Measuring Induced Variance and Consistency
- 4.1 Ground-Truth Factor Induced Variance
- 4.2 Local Directions of Variance
- 4.3 Consistency of Encodings
- 5 Experimental Setup
- 5.1 Datasets
- 5.2 Models
- 6 Results
- 6.1 The Effect of Different Per-Factor Contributions
- 6.2 The Effect of Non-global Variance Structure in the Data
- 6.3 The Effect of Non-global Variance Structure in the Models
- 7 Conclusions
- References
- Neural Topic Models for Hierarchical Topic Detection and Visualization
- 1 Introduction
- 2 Visual and Hierarchical Neural Topic Model
- 2.1 Generative Model
- 2.2 Parameterizing Path Distribution and Level Distribution
- 2.3 Parameterizing Word Distribution
- 2.4 Visualizing the Topic Tree
- 2.5 Dynamically Growing the Topic Tree
- 2.6 Autoencoding Variational Inference
- 3 Experiments
- 3.1 Tree-Structure and Visualization Quantitative Evaluation
- 3.2 Topic Coherence and Running Time Comparison
- 3.3 Visualization Qualitative Evaluation
- 4 Related Work
- 5 Conclusion
- References
- Semi-structured Document Annotation Using Entity and Relation Types
- 1 Introduction
- 2 Related Work
- 3 Problem Statement
- 4 Proposed Approach
- 4.1 Document Structure Recovery Using Generative PGM
- 4.2 Document Structure Annotation Using PLP
- 4.3 Entity and Relation Discovery
- 5 Experiments
- 6 Conclusions
- References
- Learning Disentangled Representations with the Wasserstein Autoencoder
- 1 Introduction
- 2 Importance of Total Correlation in Disentanglement
- 2.1 Total Correlation
- 2.2 Total Correlation in ELBO
- 3 Is WAE Naturally Good at Disentangling?
- 3.1 WAE
- 3.2 TCWAE
- 3.3 Estimators
- 4 Experiments
- 4.1 Quantitative Analysis: Disentanglement on Toy Data Sets
- 4.2 Qualitative Analysis: Disentanglement on Real-World Data Sets
- 5 Conclusion
- References
- Search and optimization
- Which Minimizer Does My Neural Network Converge To?
- 1 Introduction
- 2 Background
- 3 Impact of Initialization
- 4 Impact of Adaptive Optimization
- 5 Impact of Stochastic Optimization
- 6 Beyond Strong Overparameterization
- 7 Experiments
- 8 Related Work
- 9 Discussion
- References
- Information Interaction Profile of Choice Adoption
- 1 Introduction
- 1.1 Contributions
- 2 Related Work
- 3 InterRate
- 3.1 Problem Definition
- 3.2 Likelihood
- 3.3 Proof of Convexity
- 4 Experimental Setup
- 4.1 Kernel Choice
- 4.2 Parameters Learning
- 4.3 Background Noise in the Data
- 4.4 Evaluation Criteria
- 4.5 Baselines
- 5 Results
- 5.1 Synthetic Data
- 5.2 Real Data
- 6 Discussion
- 7 Conclusion
- References
- Joslim: Joint Widths and Weights Optimization for Slimmable Neural Networks
- 1 Introduction
- 2 Related Work
- 2.1 Slimmable Neural Networks
- 2.2 Neural Architecture Search
- 2.3 Channel Pruning
- 3 Methodology
- 3.1 Problem Formulation
- 3.2 Proposed Approach: Joslim
- 3.3 Relation to Existing Approaches
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Performance Gains Introduced by Joslim
- 4.3 Ablation Studies
- 5 Conclusion
- References
- A Variance Controlled Stochastic Method with Biased Estimation for Faster Non-convex Optimization
- 1 Introduction
- 2 Preliminaries
- 3 Variance Controlled SVRG with a Combined Unbiased/Biased Estimation
- 3.1 Weighted Unbiased Estimator Analysis
- 3.2 Biased Estimator Analysis
- 3.3 Convergence Analysis for Smooth Non-convex Optimization
- 3.4 Scaling Batch Samples
- 3.5 Best of Two Worlds
- 4 Application
- 5 Discussion
- References
- Very Fast Streaming Submodular Function Maximization
- 1 Introduction
- 2 Related Work
- 3 The Three Sieves Algorithm
- 4 Experimental Evaluation
- 4.1 Batch Experiments
- 4.2 Streaming Experiments
- 5 Conclusion
- References
- Dep-L0: Improving L0-Based Network Sparsification via Dependency Modeling
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Sparse Structure Learning
- 3.2 Group Sparsity
- 3.3 Gate Partition
- 3.4 Neural Dependency Modeling
- 4 Experiments
- 4.1 CIFAR10 Results
- 4.2 CIFAR100 Results
- 4.3 ImageNet Results
- 4.4 Study of Learned Sparse Structures
- 4.5 Run-Time Comparison
- 5 Conclusion and Future Work
- References
- Variance Reduced Stochastic Proximal Algorithm for AUC Maximization
- 1 Introduction
- 2 AUC Formulation
- 3 Method
- 4 Convergence Analysis
- 4.1 Bounding the Variance
- 4.2 Proof of Theorem 1
- 4.3 Complexity Analysis
- 5 Experiment
- 5.1 VRSPAM Has Lower Variance
- 5.2 VRSPAM Has Faster Convergence
- 6 Conclusion
- References
- Robust Regression via Model Based Methods
- 1 Introduction
- 2 Related Work
- 3 Robust Regression and Applications
- 4 Robust Regression via MBO
- 5 Stochastic Alternating Direction Method of Multipliers
- 5.1 SADM
- 5.2 Inner ADMM
- 5.3 Convergence
- 6 Experiments
- 6.1 Time and Objective Performance Comparison
- 6.2 Robustness Analysis
- 6.3 Classification Performance
- 7 Conclusion
- References
- Black-Box Optimizer with Stochastic Implicit Natural Gradient
- 1 Introduction
- 2 Notation and Symbols
- 3 Implicit Natural Gradient Optimization
- 3.1 Optimization with Exponential-Family Sampling
- 3.2 Implicit Natural Gradient
- 4 Update Rule for Gaussian Sampling
- 4.1 Stochastic Update
- 4.2 Direct Update for and
- 5 Convergence Rate
- 6 Optimization for Discrete Variable
- 7 Empirical Study
- 7.1 Evaluation on Synthetic Continuous Test Benchmarks
- 7.2 Evaluation on RL Test Problems
- 7.3 Evaluation on Discrete Test Problems
- 8 Conclusions
- References
- More General and Effective Model Compression via an Additive Combination of Compressions
- 1 Introduction
- 2 Related Work
- 3 Compression via an Additive Combination as Constrained Optimization
- 4 Optimization via a Learning-Compression Algorithm
- 4.1 Exactly Solvable C Step
- 5 Experiments on CIFAR10
- 5.1 Q+P: Quantization Plus Pruning
- 5.2 Q+L: Quantization Plus Low-Rank
- 5.3 L+P: Low-Rank Plus Pruning
- 6 Experiments on ImageNet
- 7 Conclusion
- References
- Hyper-parameter Optimization for Latent Spaces
- 1 Introduction
- 2 Background and Related Work
- 3 Hyper-parameter Optimization for Latent Spaces in Recommender Systems
- 3.1 The Nelder-Mead Approach
- 4 Empirical Evaluation
- 4.1 Baselines and Evaluation Protocol
- 4.2 Experiments on Real-World Data
- 4.3 Experiments on Synthetic Data
- 5 Conclusions
- References
- Bayesian Optimization with a Prior for the Optimum
- 1 Introduction
- 2 Background
- 2.1 Bayesian Optimization
- 2.2 Tree-Structured Parzen Estimator
- 3 BO with a Prior for the Optimum
- 3.1 BOPrO Priors
- 3.2 Model
- 3.3 Pseudo-posterior
- 3.4 Model and Pseudo-posterior Visualization
- 3.5 Acquisition Function
- 3.6 Putting It All Together
- 4 Experiments
- 4.1 Prior Forgetting
- 4.2 Comparison Against Strong Baselines
- 4.3 The Spatial Use-Case
- 5 Related Work
- 6 Conclusions and Future Work
- A Prior Forgetting Supplementary Experiments
- B Mathematical Derivations
- B.1 EI Derivation
- B.2 Proof of Proposition1
- C Experimental Setup
- D Spatial Real-World Application
- E Multivariate Prior Comparison
- F Misleading Prior Comparison
- G Comparison to Other Baselines
- H Prior Baselines Comparison
- I -Sensitivity Study
- J -Sensitivity Study
- References
- Rank Aggregation for Non-stationary Data Streams
- 1 Introduction
- 2 Preliminaries and Notation
- 2.1 Modeling Evolving Preferences: Evolving Mallows Model
- 3 Unbalanced Borda for Stream Ranking
- 3.1 Sample Complexity for Returning 0 on Average
- 3.2 Sample Complexity for Returning 0 with High Probability
- 3.3 Choosing Optimally
- 4 Generalizing Voting Rules
- 5 Experiments
- 5.1 Rank Aggregation for Dynamic Preferences with uBorda
- 5.2 Condorcet Winner in Dynamic Preferences
- 6 Conclusions
- References
- Adaptive Optimizers with Sparse Group Lasso for Neural Networks in CTR Prediction
- 1 Introduction
- 1.1 Adaptive Optimization Methods
- 1.2 Regularized Optimization Methods
- 1.3 Motivation
- 1.4 Outline of Contents
- 1.5 Notations and Technical Background
- 2 Algorithm
- 2.1 Closed-Form Solution
- 2.2 Concrete Examples
- 3 Convergence and Regret Analysis
- 4 Experiments
- 4.1 Experiment Setup
- 4.2 Adam vs. Group Adam
- 4.3 Adagrad vs. Group Adagrad
- 4.4 Discussion
- 5 Conclusion
- References
- Fast Conditional Network Compression Using Bayesian HyperNetworks
- 1 Introduction
- 2 Related Work
- 3 Preliminaries
- 3.1 Bayesian Neural Networks
- 3.2 Bayesian Compression
- 4 Conditional Compression
- 4.1 Conditional Bayesian Hypernetworks
- 4.2 Compression Contexts
- 4.3 Group Sparsity and Input-Output Group Sparsity Reparameterization
- 4.4 Training Methods
- 5 Experiments
- 5.1 Settings
- 5.2 Implementation
- 5.3 Conditioning on Output Subclasses and Input Domain
- 5.4 Conditioning on the Model Compression Rate
- 6 Conclusion
- References
- Active Learning in Gaussian Process State Space Model
- 1 Introduction
- 2 Background
- 3 Gaussian Process State-Space Model with Inputs
- 3.1 Learning and Prediction in GPSSM
- 4 Active Learning Strategies
- 4.1 Computation of Latest Mutual Information
- 4.2 Computation of Total Mutual Information
- 5 Experiments
- 5.1 Simulated Function
- 5.2 Pendulum and Cart-Pole
- 5.3 Twin-Rotor Aerodynamical System
- 6 Discussion
- References
- Ensembling Shift Detectors: An Extensive Empirical Evaluation
- 1 Introduction
- 2 Related Work
- 3 Shift Detectors
- 3.1 Notations and Problem Setup
- 3.2 Feature-Based Detection
- 3.3 Prediction-Based Detection
- 3.4 Limitations of Shift Detectors
- 3.5 Dataset Adaptive Significance Level
- 3.6 Detectors Ensembles
- 4 Experimental Setup
- 4.1 Datasets and Shift Simulation
- 4.2 Experiments
- 4.3 Metrics
- 4.4 Statistical Comparison
- 5 Results
- 5.1 Ensembling Shift Detectors
- 5.2 Comparison of Base Shift Detectors
- 5.3 Impact of Dataset-Adaptive Significance Level
- 5.4 Impact of Model Quality on BBSDs
- 6 Conclusion
- References
- Supervised Learning
- Adaptive Learning Rate and Momentum for Training Deep Neural Networks
- 1 Introduction
- 2 Related Work
- 3 The CGQ Method
- 3.1 Overview
- 3.2 Dynamically Adjusting the Learning Rate
- 3.3 Dynamically Adjusting Momentum
- 3.4 Optimizations for Large Datasets
- 4 Experiments
- 4.1 Ablation Test
- 4.2 Multi-class Classification on Image Datasets
- 5 Conclusion
- References
- Attack Transferability Characterization for Adversarially Robust Multi-label Classification
- 1 Introduction
- 2 Related Work
- 3 Vulnerability Assessment of Multi-label Classifiers
- 3.1 Information-Theoretic Adversarial Risk Bound
- 4 Transferrability Regularization for Adversarially Robust Multi-label Classification
- 4.1 Soft Attackability Estimator (SAE)
- 4.2 SAE Regularized Multi-label Learning
- 5 Experiments
- 5.1 Experimental Setup
- 5.2 Effectivity of Soft Attackability Estimator (SAE)
- 5.3 Effectiveness Evaluation of ARM-SAE
- 5.4 Validation of Trade-Off Between Generalization Performance on Clean Data and Adversarial Robustness
- 6 Conclusion
- References
- Differentiable Feature Selection, A Reparameterization Approach
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Preliminaries: logitNormal Law on [0, 1]
- 3.2 Parameterizing logitNormal Variables for Feature Selection
- 3.3 Sparsity Constraint: 0-Relaxation
- 3.4 Reconstruction for Feature Selection
- 4 Experiments
- 4.1 Experimental and Implementation Details
- 4.2 Independent vs Correlated Sampling Scheme
- 4.3 Feature Selection and Reconstruction
- 4.4 Quality of the Selected Features: MNIST Classification
- 4.5 Extension: cGAN
- 5 Conclusion
- References
- ATOM: Robustifying Out-of-Distribution Detection Using Outlier Mining
- 1 Introduction
- 2 Preliminaries
- 3 Method
- 3.1 ATOM: Adversarial Training with Informative Outlier Mining
- 4 Experiments
- 4.1 Setup
- 4.2 Results
- 5 Theoretical Analysis
- 5.1 Setup
- 5.2 Learning with Informative Auxiliary Data
- 5.3 Learning with Informative Outlier Mining
- 6 Related Work
- 7 Conclusion
- References
- Robust Selection Stability Estimation in Correlated Spaces
- 1 Introduction
- 2 Related Work
- 3 Feature Importance
- 4 Limits of Existing Stability Measures
- 5 Stability as Maximal Shared Importance
- 6 Experiments
- 6.1 Simulated Data
- 6.2 Stability of Standard Selection Methods
- 7 Conclusion
- References
- Gradient-Based Label Binning in Multi-label Classification
- 1 Introduction
- 2 Preliminaries
- 2.1 Multi-label Classification
- 2.2 Multivariate Gradient Boosting
- 2.3 Ensembles of Multi-label Rules
- 3 Gradient-Based Label Binning
- 3.1 Complexity Analysis
- 3.2 Mapping Labels to Bins
- 3.3 Equal-Width Label Binning
- 3.4 Aggregation of Gradients and Hessians
- 4 Evaluation
- 5 Conclusion
- References
- Joint Geometric and Topological Analysis of Hierarchical Datasets
- 1 Introduction
- 2 Problem Formulation
- 3 Background
- 3.1 Multiple Manifold Learning and Diffusion Operators
- 3.2 Simplicial Complexes and Persistent Homology
- 4 Proposed Method
- 4.1 The Sample Diffusion Operator
- 4.2 The Dataset Simplicial Complex
- 4.3 Topological Distance Between Datasets
- 5 Simulation Study
- 6 Application to HSI
- References
- Reparameterized Sampling for Generative Adversarial Networks
- 1 Introduction
- 2 Related Work
- 3 Background
- 4 The Proposed REP-GAN
- 4.1 From Independent Proposal to Dependent Proposal
- 4.2 A Tractable Structured Dependent Proposal with Reparameterized Markov Chains
- 4.3 A Practical Implementation
- 4.4 Extension to WGAN
- 5 Experiments
- 5.1 Manifold Dataset
- 5.2 Multi-modal Dataset
- 5.3 Real-World Image Dataset
- 5.4 Algorithmic Analysis
- 6 Conclusion
- References
- Asymptotic Statistical Analysis of Sparse Group LASSO via Approximate Message Passing
- 1 Introduction
- 2 Algorithm
- 2.1 Approximate Message Passing
- 2.2 Proximal Operator and Derivative
- 3 Main Result
- 3.1 State Evolution and Calibration
- 3.2 AMP Characterizes SGL Estimate
- 3.3 TPP and FDP Trade-Off of SGL
- 4 Simulation
- 4.1 State Evolution Characterization
- 4.2 Benefits of Groups
- 4.3 Extensions of SGL AMP
- 5 Discussion and Future Work
- References
- Sparse Information Filter for Fast Gaussian Process Regression
- 1 Introduction
- 2 Gaussian Process Regression
- 2.1 Full Gaussian Processes
- 2.2 Sparse Inducing Points Gaussian Processes
- 2.3 Stochastic Variational Gaussian Processes
- 2.4 Recursively Estimated Sparse Gaussian Processes
- 3 Information Filter for Sparse Gaussian Processes
- 3.1 Stochastic Hyperparameter Optimization
- 4 Experiments
- 4.1 Toy Data
- 4.2 Synthetic Data
- 4.3 Real Data
- 5 Conclusion
- References
- Bayesian Crowdsourcing with Constraints
- 1 Introduction
- 2 Problem Formulation and Preliminaries
- 2.1 Prior Works
- 3 Variational Inference for Crowdsourcing
- 3.1 Variational Bayes
- 3.2 Variational EM for Crowdsourcing
- 4 Constrained Crowdsourcing
- 4.1 Performance Analysis
- 4.2 Choosing .
- 4.3 Selecting Instance-Level Constraints
- 5 Numerical Tests
- 6 Conclusions
- References
- Text Mining and Natural Language Processing
- VOGUE: Answer Verbalization Through Multi-Task Learning
- 1 Introduction
- 2 Related Work
- 3 Task Definition
- 4 Approach
- 4.1 Dual Encoder
- 4.2 Similarity Threshold
- 4.3 Cross Attention
- 4.4 Hybrid Decoder
- 4.5 Learning
- 5 Experiments
- 5.1 Experimental Setup
- 5.2 Results
- 5.3 Ablation Study
- 5.4 Error Analysis
- 6 Conclusions
- References
- NA-Aware Machine Reading Comprehension for Document-Level Relation Extraction
- 1 Introduction
- 2 Task Formulation
- 3 NA-aware MRC (NARC)
- 3.1 Query-Context Encoder
- 3.2 Answer Vector Assembler
- 3.3 NA-Aware Predictor
- 4 Experiments
- 4.1 Dataset
- 4.2 Implementation Details
- 4.3 Baselines
- 4.4 Performance Comparison
- 4.5 Performance Analysis
- 4.6 Computational Cost Analysis
- 5 Related Work
- 6 Conclusion
- References
- Follow Your Path: A Progressive Method for Knowledge Distillation
- 1 Introduction
- 2 Related Work
- 3 Methodology
- 3.1 Background on Knowledge Distillation
- 3.2 Knowledge Distillation with Dynamic Target
- 3.3 Progressive Knowledge Teaching
- 4 Experiments
- 4.1 Setup
- 4.2 Results
- 4.3 Discussion
- 5 Conclusion
- A Appendix
- A.1 Experimental Details for Text Classification
- A.2 Full Comparison of KD in Image Recognition Sec Experiment Results of Homogeneous Architecture KD in Image Recognition
- A.3 ProKT as Approximate Mirror Descent
- References
- TaxoRef: Embeddings Evaluation for AI-driven Taxonomy Refinement
- 1 Introduction and Motivation
- 2 The Significance of Analysing Job Ads
- 3 Preliminaries and Related Work
- 4 The TaxoRef Approach
- 4.1 Task 1: Hierarchical Semantic Similarity (HSS)
- 4.2 Task 2: Generation of Embeddings
- 4.3 Task 3: Embeddings Evaluation
- 4.4 Task 4: Taxonomy Refinement
- 4.5 Benchmarking HSS
- 5 Experimental Results on 2M+ UK Online Job Ads
- 5.1 Embeddings Evaluation
- 5.2 Result Comments
- 6 Conclusion and Future Work
- References
- MaxVA: Fast Adaptation of Step Sizes by Maximizing Observed Variance of Gradients
- 1 Introduction
- 2 Preliminary and Definitions
- 3 Maximizing the Variance of Running Estimations
- 4 Experiments on Synthetic Data
- 4.1 Convergence with Stochastic Gradients
- 4.2 Convergence in the Noisy Quadratic Model
- 5 Experiments on Practical Datasets
- 5.1 Image Classification
- 5.2 Neural Machine Translation
- 5.3 General Language Understanding Evaluation (GLUE)
- 5.4 Large-Batch Pretraining for BERT
- 6 Related Work
- 7 Conclusion
- References
- Augmenting Open-Domain Event Detection with Synthetic Data from GPT-2
- 1 Introduction
- 2 Model
- 2.1 Data Generation
- 2.2 Task Modeling
- 3 Experiments
- 3.1 Datasets and Hyper-parameters
- 3.2 Baselines
- 3.3 Results
- 3.4 Ablation Study
- 3.5 Analysis
- 4 Related Work
- 5 Conclusion
- References
- Enhancing Summarization with Text Classification via Topic Consistency
- 1 Introduction
- 2 Related Work
- 2.1 Abstractive Summarization
- 2.2 Combining Summarization and Text Classification
- 2.3 Summarization Evaluation Metrics
- 3 Methodology
- 3.1 XMTC for Topic Distribution Prediction
- 3.2 Enhancing Summarization with Topic Consistency
- 3.3 Topic Consistency as an Additional Evaluation Metric
- 4 Experiments
- 4.1 Datasets
- 4.2 Comparing Methods
- 4.3 Main Results
- 4.4 Human Evaluation
- 5 Conclusions
- References
- Transformers: ``The End of History'' for Natural Language Processing?
- 1 Introduction
- 2 Related Work
- 3 Tasks
- 3.1 Propaganda Detection
- 3.2 Keyphrase Extraction
- 4 Method
- 4.1 Token Classification
- 4.2 Sequence Classification
- 5 Experimental Setup
- 5.1 Data
- 5.2 Parameter Setting
- 6 Experiments and Results
- 6.1 Token Classification
- 6.2 Sequence Classification
- 7 Discussion
- 8 Conclusion and Future Work
- References
- Image Processing, Computer Vision and Visual Analytics
- Subspace Clustering Based Analysis of Neural Networks
- 1 Introduction
- 2 Background and Method
- 2.1 Sparse Spectral Clustering
- 2.2 Centered Kernel Alignment
- 2.3 Meta Algorithm
- 3 Problem and Experimental Setup
- 3.1 Problem Formulation
- 3.2 Experimental Details
- 4 Analysis of Network Training Dynamics
- 4.1 Community Structure via Graph Modularity
- 4.2 Layer-Wise Training Dynamics and Convergence
- 4.3 Comparison with Linear-CKA
- 5 Analysis of Network Architecture
- 5.1 Observing the Effects of Depth
- 5.2 Observing the Effects of Width
- 5.3 Observing the Effects of Epochs
- 5.4 Observing the Effects of Quantity of Training Data
- 6 Analysis of Inputs
- 6.1 Layer-Wise Latent Space Visualisation
- 6.2 Model Explanation by Instance Neighbourhood Visualisation
- 7 Conclusions
- References
- Invertible Manifold Learning for Dimension Reduction
- 1 Introduction
- 2 Related Work
- 3 Proposed Method
- 3.1 Problem Statement
- 3.2 Methods for Structure Preservation
- 3.3 Linear Compression
- 3.4 Network Implementation
- 4 Experiment
- 4.1 Methods Comparison
- 4.2 Latent Space Interpolation
- 4.3 Analysis
- 5 Conclusion
- References
- Small-Vote Sample Selection for Label-Noise Learning
- 1 Introduction
- 2 Related Work
- 3 Proposed Method
- 3.1 Preliminaries
- 3.2 Hierarchical Voting Scheme (HVS)
- 3.3 Adaptive Clean Data Rate Estimation Strategy (ACES)
- 4 Experiments
- 4.1 Comparisons with State-of-the-Arts
- 4.2 Ablation Studies
- 5 Conclusion
- References
- Iterated Matrix Reordering
- 1 Introduction
- 2 Related Work
- 3 Iterative Ordering Using Convolution
- 4 Experiments
- 4.1 Experimental Setup
- 4.2 Datasets
- 4.3 Results
- 5 Conclusions and Future Work
- References
- Semi-supervised Semantic Visualization for Networked Documents
- 1 Introduction
- 2 Related Work
- 3 Model Architecture and Analysis
- 3.1 Coordinate-Based Distribution
- 3.2 Label Modeling
- 3.3 Link and Content Modeling
- 3.4 The Complete Model
- 4 Experiments
- 4.1 Quantitative Evaluation
- 4.2 Visualization
- 4.3 User Study
- 5 Conclusion
- References
- Self-supervised Multi-task Representation Learning for Sequential Medical Images
- 1 Introduction
- 2 Related Works
- 2.1 Self-supervised Learning
- 2.2 Multi-task Learning
- 3 Method
- 3.1 Self-supervised Multi-tasking Learning
- 3.2 Integration with Instance Discrimination
- 4 Experiments
- 4.1 Datasets
- 4.2 Experimental Setup
- 4.3 Results
- 5 Conclusion
- References
- Label-Assisted Memory Autoencoder for Unsupervised Out-of-Distribution Detection
- 1 Introduction
- 2 Background
- 2.1 AE-based Detectors
- 2.2 Non-AE-Based Detectors
- 3 Label-Assisted Memory AutoEncoder
- 3.1 Label-Assisted Memory AutoEncoder (LAMAE)
- 3.2 LAMAE with Complexity Normalizer (LAMAE+)
- 4 Experimental Studies
- 4.1 Experimental Setup
- 4.2 Experiment 1: Comparative Studies with SOTA Detectors
- 4.3 Experiment 2: Analysis of LAMAE+
- 5 Conclusion
- References
- Quantized Gromov-Wasserstein
- 1 Introduction
- 2 Quantized Gromov-Wasserstein
- 3 Theoretical Error Bounds
- 4 Experiments
- 5 Discussion
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.