Machine Learning and Knowledge Discovery in Databases. Research Track

Name: Machine Learning and Knowledge Discovery in Databases. Research Track | European Conference, ECML PKDD 2021, Bilbao, Spain, September 13-17, 2021, Proceedings, Part II
Brand: Springer
Price: 106.99 EUR
Availability: OnlineOnly

European Conference, ECML PKDD 2021, Bilbao, Spain, September 13-17, 2021, Proceedings, Part II

Nuria Oliver Fernando Pérez-Cruz Stefan Kramer Jesse Read Jose A. Lozano(Editor)

Springer (Publisher)

Published on 9. September 2021

XXXV, 817 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-86520-7 (ISBN)

€106.99incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.

Alles über E-Books, Kopierschutz & 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

Content

Intro
Preface
Organization
Contents - Part II
Generative Models
Non-exhaustive Learning Using Gaussian Mixture Generative Adversarial Networks
1 Introduction
2 Related Work
3 Background
4 Methodology
5 Experiments
6 Conclusion
References
Unsupervised Learning of Joint Embeddings for Node Representation and Community Detection
1 Introduction
2 Related Work
2.1 Community Detection
2.2 Node Representation Learning
2.3 Joint Community Detection and Node Representation Learning
3 Methodology
3.1 Problem Formulation
3.2 Variational Model
3.3 Design Choices
3.4 Practical Aspects
3.5 Complexity
4 Experiments
4.1 Synthetic Example
4.2 Datasets
4.3 Baselines
4.4 Settings
4.5 Discussion of Results
4.6 Hyperparameter Sensitivity
4.7 Training Time
4.8 Visualization
5 Conclusion
References
GraphAnoGAN: Detecting Anomalous Snapshots from Attributed Graphs
1 Introduction
2 Related Work
3 Problem Definition
4 Proposed Algorithm
4.1 GAN Modeling
4.2 Architecture
4.3 Training Procedure
5 Datasets
6 Experiments
6.1 Baselines
6.2 Comparative Evaluation
6.3 Side-by-Side Diagnostics
7 Conclusion
References
The Bures Metric for Generative Adversarial Networks
1 Introduction
2 Method
3 Empirical Evaluation of Mode Collapse
3.1 Artificial Data
3.2 Real Images
4 High Quality Generation Using a ResNet Architecture
5 Conclusion
References
Generative Max-Mahalanobis Classifiers for Image Classification, Generation and More
1 Introduction
2 Background and Related Work
2.1 Energy-Based Models
2.2 Alternatives to the Softmax Classifier
3 Methodology
3.1 Approach 1: Discriminative Training
3.2 Approach 2: Generative Training
3.3 Approach 3: Joint Training
3.4 GMMC for Inference
4 Experiments
4.1 Hybrid Modeling
4.2 Calibration
4.3 Out-Of-Distribution Detection
4.4 Robustness
4.5 Training Stability
4.6 Joint Training
5 Conclusion and Future Work
References
Gaussian Process Encoders: VAEs with Reliable Latent-Space Uncertainty
1 Introduction
1.1 Contributions
2 Background
2.1 Variational Autoencoder
2.2 Latent Variance Estimates of NN
2.3 Mismatch Between the Prior and Approximate Posterior
3 Methodology
3.1 Gaussian Process Encoder
3.2 The Implications of a Gaussian Process Encoder
3.3 Out-of-Distribution Detection
4 Experiments
4.1 Log Likelihood
4.2 Uncertainty in the Latent Space
4.3 Benchmarking OOD Detection
4.4 OOD Polution of the Training Data
4.5 Synthesizing Variants of Input Data
4.6 Interpretable Kernels
5 Related Work
6 Conclusion
References
Variational Hyper-encoding Networks
1 Introduction
2 Variational Autoencoder (VAE)
3 Variational Hyper-encoding Networks
3.1 Hyper-auto-encoding Problem
3.2 Hyper-encoding Problem
3.3 Minimum Description Length
3.4 Compact Hyper-decoder Architecture
3.5 Applications
4 Experiments
4.1 Data Sets
4.2 Model Settings
4.3 Model Behavior
4.4 Robust Outlier Detection
4.5 Novelty Discovery
5 Related Work
6 Conclusion
References
Principled Interpolation in Normalizing Flows
1 Introduction
2 An Intuitive Solution
3 Normalizing Flows
4 Base Distributions on p-Norm Spheres
4.1 The Case p = 1
4.2 The Case p = 2
5 Experiments
5.1 Performance Metrics and Setup
5.2 Data
5.3 Architecture
5.4 Quantitative Results
5.5 Qualitative Results
6 Related Work
7 Conclusion
References
CycleGAN Through the Lens of (Dynamical) Optimal Transport
1 Introduction
2 Desiderata for UDT and Analysis of CycleGAN
2.1 What Should Be the Properties of a UDT Solution?
2.2 CycleGAN Is Biased Towards Low Energy Transformations
3 UDT as Optimal Transport
3.1 A (Dynamical) OT Model for UDT
3.2 Regularity of OT Maps
3.3 Computing the Inverse
4 A Residual Instantiation from Dynamical OT
4.1 Linking the Dynamical Formulation with CycleGAN
4.2 A Typical UDT Task
4.3 Imbalanced CelebA task
5 Related Work
6 Discussion and Conclusion
References
Decoupling Sparsity and Smoothness in Dirichlet Belief Networks
1 Introduction
2 Preliminary Knowledge
3 Sparse and Smooth Dirichlet Belief Networks
3.1 Generative Process
3.2 Necessity of Fixing biK(l)=1
4 Related Work
5 ssDirBN for Relational Modelling
5.1 Inference
5.2 Experimental Results
6 Conclusion
References
Algorithms and Learning Theory
Self-bounding Majority Vote Learning Algorithms by the Direct Minimization of a Tight PAC-Bayesian C-Bound
1 Introduction
2 Majority Vote Learning
2.1 Notations and Setting
2.2 Gibbs Risk, Joint Error and C-Bound
2.3 Related Works
3 PAC-Bayesian C-Bounds
3.1 An Intuitive Bound-McAllester's View
3.2 A Tighter Bound-Seeger's View
3.3 Another Tighter Bound-Lacasse's View
4 Self-bounding Algorithms for PAC-Bayesian C-Bounds
4.1 Algorithm Based on McAllester's View
4.2 Algorithm Based on Seeger's View
4.3 Algorithm Based on Lacasse's View
5 Experimental Evaluation
5.1 Empirical Setting
5.2 Analysis of the Results
6 Conclusion and Future Work
References
Midpoint Regularization: From High Uncertainty Training Labels to Conservative Classification Decisions
1 Introduction
2 Related Work
3 Label Smoothing over Midpoint Samples
3.1 Preliminaries
3.2 Midpoint Generation
3.3 Learning Smoothing Distribution for Midpoints
3.4 Optimization
4 Experiments
4.1 Datasets, Baselines, and Settings
4.2 Predictive Accuracy
4.3 Ablation Studies
4.4 Uncertainty Label and Conservative Classification
4.5 Testing on Out-of-Distribution Data
5 Conclusion and Future Work
References
Learning Weakly Convex Sets in Metric Spaces
1 Introduction
2 Preliminaries
3 Weak Convexity in Metric Spaces
3.1 Some Basic Properties of Weakly Convex Sets
4 Learning in the Extensional Problem Setting
4.1 Application Scenario: Vertex Classification
5 The Intensional Problem Setting
5.1 Learning Weakly Convex Boolean Functions
5.2 Learning Weakly Convex Axis-Aligned Hyperrectangles
6 Concluding Remarks
References
Disparity Between Batches as a Signal for Early Stopping
1 Introduction
2 Related Work
3 Generalization Penalty
4 Gradient Disparity
5 Early Stopping Criterion
6 Discussion and Final Remarks
References
Learning from Noisy Similar and Dissimilar Data
1 Introduction
2 Problem Setup
3 Loss Correction Approach
4 Weighted Classification Approach
5 Experiments
6 Conclusion and Future Work
References
Knowledge Distillation with Distribution Mismatch
1 Introduction
2 Related Works
3 Framework
3.1 Problem Definition
3.2 Proposed Method KDDM
4 Experiments and Discussions
4.1 Datasets
4.2 Baselines
4.3 Results on MNIST
4.4 Results on CIFAR-10
4.5 Results on CIFAR-100
4.6 Distillation When Teacher-Data and Student-Data Are Identical
5 Conclusion
References
Certification of Model Robustness in Active Class Selection
1 Introduction
1.1 Active Class Selection Constitutes a Domain Gap
1.2 A Qualitative Intuition from Information Theory
2 A Quantitative Perspective from Learning Theory
2.1 Quantification of the Domain Gap
2.2 Certification of Domain Robustness for Binary Predictors
3 Experiments
3.1 Binary (, ) Certificates Are Tight
3.2 Binary (, ) Certificates in Astro-Particle Physics
4 Related Work
5 Conclusion
References
Graphs and Networks
Inter-domain Multi-relational Link Prediction
1 Introduction
2 Preliminary
2.1 RESCAL
2.2 Optimal Transport
2.3 Maximum Mean Discrepancy
3 Problem Setting and Proposed Method
3.1 Problem Setting
3.2 Proposed Objective Function
4 Experiments
4.1 Datasets
4.2 Evaluation Methods and Baselines
4.3 Implementation Details
4.4 Experimental Results
5 Related Work
6 Conclusion and Future Work
References
GraphSVX: Shapley Value Explanations for Graph Neural Networks
1 Introduction
2 Related Work
3 Preliminary Concepts and Background
3.1 Graph Neural Networks
3.2 The Shapley Value
4 A Unified Framework for GNN Explainers
5 Proposed Method
5.1 Mask and Graph Generators
5.2 Explanation Generator
5.3 Decomposition Model
5.4 Efficient Approximation Specific to GNNs
5.5 Desirable Properties of Explanations
6 Experimental Evaluation
6.1 Synthetic and Real Datasets with Ground Truth
6.2 Real-World Datasets Without Ground Truth
7 Conclusion
References
Multi-view Self-supervised Heterogeneous Graph Embedding
1 Introduction
2 Related Work
2.1 Self-supervised Learning on Graphs
2.2 Heterogeneous Graph Embedding
3 The Proposed Model
3.1 Model Framework
3.2 Heterogeneous Context Encoding
3.3 Multi-view Contrastive Learning
4 Experiment
4.1 Experimental Setup
4.2 Node Classification
4.3 Link Prediction
4.4 Ablation Study
4.5 Visualization
5 Conclusion
References
Semantic-Specific Hierarchical Alignment Network for Heterogeneous Graph Adaptation
1 Introduction
2 Related Work
2.1 Heterogeneous Graph Neural Network
2.2 Graph Domain Adaptation
3 Preliminaries
4 The Proposed Model
4.1 Semantic-Specific GNN for DA
4.2 Hierarchical Domain Alignment
4.3 Optimization Objective
4.4 Discussion of the Proposed Model
5 Experiments
5.1 Datasets
5.2 Baselines and Implementation Details
5.3 Results
5.4 Analysis
6 Conclusion
References
The KL-Divergence Between a Graph Model and its Fair I-Projection as a Fairness Regularizer
1 Introduction
2 Related Work
3 Fair Information Projection
3.1 Notation
3.2 Fairness Constraints
3.3 Information Projection
4 The KL-Divergence to the I-projection as a Fairness Regularizer
4.1 I-Projection Regularization
4.2 Practical Considerations
5 Experiments
5.1 Datasets
5.2 Algorithms
5.3 Fair Graph Embedding Baselines
5.4 Evaluation
5.5 Results
6 Conclusion
References
On Generalization of Graph Autoencoders with Adversarial Training
1 Introduction
2 Related Work
3 Preliminaries
3.1 Graph Autoencoders
3.2 Graph Variational Autoencoders
3.3 Adversarial Training
4 Graph Adversarial Training
4.1 Adversarial Training in Graph Autoencoder
4.2 Adversarial Training in Variational Graph Autoencoder
5 Experiments
5.1 Link Prediction
5.2 Node Clustering
5.3 Graph Anomaly Detection
6 Understanding Adversarial Training
6.1 The Impact of
6.2 The Impact of T
6.3 The Impact of
6.4 Performance w.r.t. Degree
7 Conclusion
References
Inductive Link Prediction with Interactive Structure Learning on Attributed Graph
1 Introduction
2 Model Formulation
2.1 Notations and Definitions
2.2 Overview of PaGNN
2.3 Broadcasting Operation
2.4 Aggregation Operation
2.5 Edge Representation Learning
2.6 Cache Strategy in Inference
2.7 Summary
3 Experiments
3.1 Experiment Setup
3.2 Performance Comparison
3.3 Efficiency Analysis
4 Related Work
5 Conclusion
References
Representation Learning on Multi-layered Heterogeneous Network
1 Introduction
2 Related Work
3 Definitions and Problem Formulation
4 Model Architecture
4.1 Intra-layer Proximity Modeling
4.2 Cross-Layer Proximity Modeling
4.3 Learning Strategy
5 Experiments
5.1 Setup
5.2 Node Classification
5.3 Node Clustering
5.4 Link Prediction
5.5 Network Visualization
5.6 Model Analysis
5.7 Case Study
6 Conclusion
References
Adaptive Node Embedding Propagation for Semi-supervised Classification
1 Introduction
2 Preliminaries
2.1 Problem Definition
2.2 Graph Convolutional Networks
3 Our Approach
3.1 Embedding Propagation Loss
3.2 Anti-smoothness Loss
3.3 Adaptive Propagation Control
3.4 Architecture of ANEPN
4 Experiments
4.1 Settings
4.2 Results
5 Related Work
5.1 Graph-Based Semi-supervised Learning
5.2 Graph Convolutional Networks
6 Conclusion
References
Probing Negative Sampling for Contrastive Learning to Learn Graph Representations
1 Introduction
2 Related Work
2.1 Graph Representation Learning
2.2 Contrastive Learning and Negative Sampling
3 Preliminaries and Problem Formulation
3.1 Determinantal Point Process
3.2 Notations and Definitions
3.3 Problem Setup
4 The Proposed Approach
4.1 Graph Embeddings
4.2 Resolving Class Collision
4.3 Sampling Diverse Negative Examples
4.4 Node-Wise Contrastive Learning Loss
5 Experimental Results
5.1 Experimental Setup
5.2 RQ1: Performance Comparison
5.3 RQ2: Ablation Study
5.4 RQ3: Parameter Analysis
5.5 RQ4: Visualization Results
6 Conclusion
References
Beyond Low-Pass Filters: Adaptive Feature Propagation on Graphs
1 Introduction
2 Preliminaries
3 Proposed Approach
4 Experiments
4.1 Experimental Setup
4.2 Results and Discussion
5 Related Work
6 Conclusion
References
Zero-Shot Scene Graph Relation Prediction Through Commonsense Knowledge Integration
1 Introduction
2 Motivating Analysis
2.1 Ignorance yet Importance of Zero-Shot Triplets
2.2 Commonsense Knowledge from ConceptNet Neighbors
2.3 Commonsense Knowledge from ConceptNet Paths
3 COACHER
3.1 Backbone Scene Graph Generation Pipeline
3.2 Commonsense Integrator
4 Experiments
4.1 Experimental Settings
4.2 Hyper-parameter Setting
4.3 Performance Evaluations
4.4 Case Studies
5 Related Work
5.1 Scene Graph Generation
5.2 External Knowledge Enhanced Deep Neural Networks
6 Conclusion
References
Graph Fraud Detection Based on Accessibility Score Distributions
1 Introduction
2 Related Work
3 Preliminaries
3.1 Random Walk with Restart
3.2 RWR for Bipartite Graphs
4 Proposed Method
4.1 Accessibility
4.2 Skewness in Accessibility Score Distributions
4.3 Theoretical Analysis
4.4 SkewA
5 Experiments
5.1 Setup
5.2 Robustness to Sparse Frauds
5.3 Camouflage-Resistance
5.4 Effects of Camouflage Ratio
5.5 Effectiveness of Theoretical Analysis
6 Conclusion
References
Correlation Clustering with Global Weight Bounds
1 Introduction
2 Related Work
3 Problem Definition
4 Theoretical Results and Algorithms
4.1 PC-Reduction
4.2 Preserving the Approximation Factor Across PC-Reduction
4.3 Ultimate Global Weight Bounds
4.4 Algorithms
5 Experiments
5.1 Analysis of the Global-Weight-Bounds Condition
5.2 Application to Fair Clustering
6 Conclusions
References
Modeling Multi-factor and Multi-faceted Preferences over Sequential Networks for Next Item Recommendation
1 Introduction
2 Related Work
3 Problem Formulation
4 The Proposed Method
4.1 Construction and Embedding of Sequential Networks
4.2 Multi-factor Preference Modeling
4.3 Multi-faceted Preference Modeling
4.4 Recommendation Model
5 Experiment
5.1 Experimental Setup
5.2 Model Comparison
5.3 Ablation Studies
5.4 Hyper-Parameter Study
6 Conclusion
References
PATHATTACK: Attacking Shortest Paths in Complex Networks
1 Introduction
2 Problem Statement
3 Proposed Method: PATHATTACK
3.1 Path Cutting as Set Cover
3.2 Linear Programming Formulation
3.3 Constraint Generation
3.4 PATHATTACK
4 Experiments
4.1 Baseline Methods
4.2 Synthetic and Real Networks
4.3 Experimental Setup
4.4 Results
5 Related Work
6 Conclusions
References
Embedding Knowledge Graphs Attentive to Positional and Centrality Qualities
1 Introduction
2 Related Work
3 Background and Notation
4 Method
4.1 Model Formulation
4.2 Centrality-Attentive Embedding
4.3 Position-Attentive Embedding:
5 Theoretical Analysis
5.1 Connection to Preceding KGE Methods
5.2 Expressive Power
5.3 Complexity Analysis
6 Experiments
6.1 Transductive Link Prediction Experiment
6.2 Inductive Link Prediction Experiment
7 Conclusion
References
Interpretation, Explainability, Transparency, Safety
Reconnaissance for Reinforcement Learning with Safety Constraints
1 Introduction
2 Background
3 Theory
4 Reconnaissance-MDP and Planning-MDP
4.1 The RP Algorithm
4.2 Threat Decomposition and the Approximate RP Algorithm
5 Experiments
5.1 Experimental Setup
5.2 Frozen Lake Results
5.3 High-Dimensional Tasks: Point Gather, Circuit, and Jam
5.4 Recycling the Threat Function for a Different Environment
6 Related Work
6.1 Optimization-Based Approaches
6.2 Sampling and Learning-Based Approaches
7 Conclusion
References
VeriDL: Integrity Verification of Outsourced Deep Learning Services
1 Introduction
2 Preliminaries
3 Problem Statement
4 Authentication Method
4.1 Setup Protocol
4.2 Certify Protocol
4.3 Verify Protocol
4.4 Dealing with Decimal and Negative Values
5 Experiments
5.1 Setup
5.2 Efficiency of VeriDL
5.3 Verification vs. Re-computation of Model Updates
5.4 Robustness of Verification
6 Related Work
7 Conclusion and Future Work
References
A Unified Batch Selection Policy for Active Metric Learning
1 Introduction
2 Related Work
2.1 Perceptual Metric Learning
2.2 Active Learning for Classification
2.3 Active Learning of Perceptual Metrics
3 Proposed Method
3.1 Triplet-Based Active Metric Learning
3.2 Joint Entropy Measure for Batch Selection
3.3 Maximum-Entropy Model of the Joint Distribution
3.4 Greedy Algorithm for Batch Selection
3.5 Recursive Computation of Determinant of Covariance Matrix
4 Experiments
5 Conclusion, Limitations, and Future Work
References
Off-Policy Differentiable Logic Reinforcement Learning
1 Introduction
2 Preliminary
2.1 First-Order Logic Programming
2.2 Differentiable Inductive Logic Programming
2.3 Maximum Entropy Reinforcement Learning
3 Off-Policy Differentiable Logic Reinforcement Learning
3.1 Differentiable Logic Policy
3.2 Approximate Inference
3.3 Off-Policy Training
3.4 Hierarchical Policy Implementation
4 Experiments
4.1 Methods for Evaluation
4.2 Experiment Setting
4.3 Results and Analysis
4.4 Rabbids Game
5 Discussion
6 Conclusion
References
Causal Explanation of Convolutional Neural Networks
1 Introduction
2 Related Work
3 Causal Explanation of CNNs
3.1 Filters as Actual Causes
4 Experiments
4.1 Evaluating Visualizations
4.2 MNIST
4.3 Weakly Supervised Object Localization (WSOL)
4.4 Fine-Tuning Parameters and Compact Representation
5 Conclusion and Future Work
References
Interpretable Counterfactual Explanations Guided by Prototypes
1 Introduction
2 Related Work
3 Methodology
3.1 Background
3.2 Prototype Loss Term
3.3 Using K-D Trees as Class Representations
3.4 Categorical Variables
3.5 Removing Lpred
3.6 FISTA Optimization
4 Experiments
4.1 Evaluation
4.2 Handwritten Digits
4.3 Breast Cancer Wisconsin (Diagnostic) Dataset
4.4 Adult (Census) Dataset
5 Discussion
References
Finding High-Value Training Data Subset Through Differentiable Convex Programming
1 Introduction
2 Related Work
3 High-Value Data Subset Selection
3.1 Motivation and Problem Formulation
3.2 Joint Online Training of Subset Selection and Learning Model
3.3 Trainable Convex Online Subset Selection Layer
4 Experiment
4.1 Experimental Setup
4.2 Valuable Training Data
4.3 Fixing Mislabelled Data
4.4 Qualitative Analogy of Data Valuation
4.5 Ranking Function for Value Points
4.6 Computational Complexity
5 Conclusion
References
Consequence-Aware Sequential Counterfactual Generation
1 Introduction
2 Related Work
3 Problem Statement
3.1 Actions, Sequences of Actions and States
3.2 Consequence-Aware Cost Model
3.3 Consequence-Aware Sequential Counterfactual Generation
4 Consequence-Aware Sequential Counterfactuals (CSCF)
5 Experiments
5.1 Sequence Costs of Sequential Counterfactuals
5.2 Diversity of Sequential Counterfactuals
5.3 Effect of the Action Positions on Achieving the Counterfactual
6 Conclusion and Future Work
References
Studying and Exploiting the Relationship Between Model Accuracy and Explanation Quality*-10pt
1 Introduction
1.1 Why It Is Important to Evaluate the Quality of Explanations?
2 Background and Related Work
2.1 Explainability in Neural Networks
2.2 Evaluating Explanation Quality
3 Definitions
4 Investigating the Relationship Between Model Accuracy and Explanation Quality
4.1 Evaluating Explanation Quality
4.2 Comparing Different Models
4.3 Studying Explanations as a Model Evolves During Training
4.4 Selecting Models Based on Explanation Quality
5 Experimental Evaluation
5.1 Data
5.2 Implementation
5.3 Results - Comparing Different Models
5.4 Results - Studying a Model at Different Iterations During Training
5.5 Using Explanations for Model Selection
6 Conclusion
References
Explainable Multiple Instance Learning with Instance Selection Randomized Trees
1 Introduction
2 Instance Selection Randomized Trees
3 Related Work
4 Experiments
4.1 Private Dataset
4.2 Public Datasets
5 Conclusion
References
Adversarial Representation Learning with Closed-Form Solvers
1 Introduction
2 Prior Work
3 Problem Setting
3.1 Motivating Exact Solvers
4 Approach
4.1 Closed-Form Adversary and Target Predictor
4.2 Optimal Embedding Dimensionality
4.3 Gradient of Closed-Form Solution
5 Experiments
5.1 Fair Classification
5.2 Mitigating Sensitive Information Leakage
5.3 Ablation Study on Mixture of Four Gaussians
6 Concluding Remarks
References
Learning Unbiased Representations via Rényi Minimization
1 Introduction
2 Related Work
3 Problem Statement
3.1 Metrics for Continuous Statistical Dependence
4 Theoretical Properties
4.1 Consistency of the HGR_NN
4.2 Theoretical Comparison Against Simple Adversarial Algorithms
5 Method
6 Experiments
6.1 Synthetic Scenario
6.2 MNIST with Continuous Color Intensity
6.3 Real-World Experiments
7 Conclusion
References
Diversity-Aware k-median: Clustering with Fair Center Representation
1 Introduction
2 Related Work
3 Problem Statement and Complexity
3.1 NP-Hardness
3.2 Fixed-Parameter Intractability
3.3 Hardness of Approximation
4 Approximable Instances
4.1 Non-intersecting Facility Groups
4.2 Two Facility Groups
4.3 The Case i ri & k
5 Proposed Methods
5.1 Local Search
5.2 Relaxing the Representation Constraints
6 Experimental Evaluation
6.1 Results
7 Conclusion
References
Sibling Regression for Generalized Linear Models
1 Introduction
2 Related Work
3 Preliminaries
4 Sibling Regression for Generalized Linear Models
4.1 Extension to GLM
5 Experiments
5.1 Synthetic Experiment
5.2 Discover Life Moth Observations
6 Ethical Impact
7 Conclusion
References
Privacy Amplification via Iteration for Shuffled and Online PNSGD
1 Introduction
2 Related Work
3 Shuffled PNSGD
4 Asymptotic Analysis for When Using Shuffling and Fixed Noises
4.1 Laplace Noise
4.2 Gaussian Noise
5 Multiple Epochs Composition
6 Online Results for Decaying Noises
6.1 Laplace Noise
6.2 Gaussian Noise
7 Conclusion
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Machine Learning and Knowledge Discovery in Databases. Research Track

Description

More details

Other editions

Additional editions

Content

System requirements