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 I
Brand: Springer
Price: 106.99 EUR
Availability: OnlineOnly

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

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

Springer (Verlag)

Erschienen am 9. September 2021

XLI, 806 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-86486-6 (ISBN)

106,99 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Invited Talks Abstracts
WuDao: Pretrain the World
The Value of Data for Personalization
AI Fairness in Practice
Safety and Robustness for Deep Learning with Provable Guarantees
Contents - Part I
Online Learning
Routine Bandits: Minimizing Regret on Recurring Problems
1 Introduction
2 The Routine Bandit Setting
3 The KLUCB-RB Strategy
4 Sketch of Proof
5 Numerical Experiments
5.1 More Arms Than Bandits: A Beneficial Case
5.2 Increasing the Number of Bandit Instances
5.3 Critical Settings
6 Conclusion
References
Conservative Online Convex Optimization
1 Introduction
2 Background
3 Problem Formulation
4 The Conservative Projection Algorithm
4.1 The Conservative Ball
4.2 Description of the CP Algorithm
4.3 Analysis of the CP Algorithm
5 Experiments
5.1 Synthetic Regression Dataset
5.2 Online Classification: The IMDB Dataset
5.3 Online Classification: The SpamBase Dataset
6 Conclusions
References
Knowledge Infused Policy Gradients with Upper Confidence Bound for Relational Bandits
1 Introduction
2 Problem Setting
3 Knowledge Infused Policy Gradients
4 Formulation of Knowledge Infusion
5 Regret Bound for KIPG
6 KIPG-Upper Confidence Bound
7 Experiments
7.1 Simulated Domains
7.2 Real-World Datasets
8 Conclusion and Future Work
References
Exploiting History Data for Nonstationary Multi-armed Bandit
1 Introduction
2 Related Works
3 Problem Formulation
4 The BR-MAB Algorithm
4.1 Break-Point Prediction Procedure
4.2 Recurrent Concepts Equivalence Test
4.3 Regret Analysis for Generic CD-MABs
4.4 Regret Analysis for the Break-Point Prediction Procedure
5 Experiments
5.1 Toy Example
5.2 Synthetic Setting
5.3 Yahoo! Setting
6 Conclusion and Future Works
References
High-Probability Kernel Alignment Regret Bounds for Online Kernel Selection
1 Introduction
1.1 Related Work
2 Problem Setting
3 A Nearly Optimal High-Probability Regret Bound
3.1 Warm-Up
3.2 A More Efficient Algorithm
3.3 Regret Bound
3.4 Time Complexity Analysis
4 Regret-Performance Trade-Off
4.1 Regret Bound
4.2 Budgeted EA2OKS
5 Experiments
5.1 Experimental Setting
5.2 Experimental Results
6 Conclusion
References
Reinforcement Learning
Periodic Intra-ensemble Knowledge Distillation for Reinforcement Learning
1 Introduction
2 Related Work
3 Background
4 Method
4.1 Overview
4.2 Ensemble Initialization
4.3 Joint Training
4.4 Intra-ensemble Knowledge Distillation
5 Experiments
5.1 Experimental Setup
5.2 Effectiveness of PIEKD
5.3 Effectiveness of Knowledge Distillation for Knowledge Sharing
5.4 Effectiveness of Selecting the Best-Performing Agent as the Teacher
5.5 Ablation Study on Ensemble Size
5.6 Ablation Study on Distillation Interval
6 Conclusion
References
Learning to Build High-Fidelity and Robust Environment Models
1 Introduction
2 Related Work
2.1 Simulator Building
2.2 Model-Based Reinforcement Learning
2.3 Offline Policy Evaluation
2.4 Robust Reinforcement Learning
3 Preliminaries
3.1 Markov Decision Process
3.2 Dual Markov Decision Process
3.3 Imitation Learning
4 Robust Learning to Simulate
4.1 Problem Definition
4.2 Single Behavior Policy Setting
4.3 Robust Policy Setting
5 Experiments
5.1 Experimental Protocol
5.2 Studied Environments and Baselines
5.3 Performance on Policy Value Difference Evaluation
5.4 Performance on Policy Ranking
5.5 Performance on Policy Improvement
5.6 Analysis on Hyperparameter
6 Conclusion
References
Ensemble and Auxiliary Tasks for Data-Efficient Deep Reinforcement Learning
1 Introduction
2 Related Works
3 Background
3.1 Markov Decision Process and RL
3.2 Rainbow Agent
4 Rainbow Ensemble
5 Auxiliary Tasks for Ensemble RL
5.1 Network Architecture
5.2 Model Learning as Auxiliary Tasks
5.3 Object and Event Based Auxiliary Tasks
6 Theoretical Analysis
7 Experiments
7.1 Comparison to Prior Works
7.2 Bias-Variance-Covariance Measurements
7.3 On Independent Training of Ensemble
7.4 The Importance of Auxiliary Tasks
7.5 On Distributing the Auxiliary Tasks
8 Conclusions
References
Multi-agent Imitation Learning with Copulas
1 Introduction
2 Preliminaries
3 Modeling Multi-agent Interaction with Copulas
3.1 Copulas
3.2 Multi-agent Imitation Learning with Copulas
4 Related Work
5 Experiments
5.1 Experimental Setup
5.2 Results
5.3 Generalization of Copula
5.4 Copula Visualization
5.5 Trajectory Generation
6 Conclusion and Future Work
A Dataset Details
B Implementation Details
References
CMIX: Deep Multi-agent Reinforcement Learning with Peak and Average Constraints
1 Introduction
2 Background
2.1 QMIX
2.2 Constrained Reinforcement Learning
3 Problem Formulation
4 CMIX
4.1 Multi-objective Constrained Problem
4.2 CMIX Architecture
4.3 Gap Loss Function
4.4 CMIX Algorithm
5 Experiments
5.1 Blocker Game with Travel Cost
5.2 Vehicular Network Routing Optimization
5.3 Gap Loss Coefficient
6 Related Work
7 Conclusion
References
Model-Based Offline Policy Optimization with Distribution Correcting Regularization
1 Introduction
2 Preliminary
2.1 Markov Decision Processes
2.2 Offline RL
2.3 Model-Based RL
3 A Lower Bound of the True Expected Return
4 Method
4.1 Overall Framework
4.2 Ratio Estimation via DICE
5 Experiment
5.1 Comparative Evaluation
5.2 Empirical Analysis
6 Related Work
6.1 Model-Free Offline RL
6.2 Model-Based Offline RL
7 Conclusion
References
Disagreement Options: Task Adaptation Through Temporally Extended Actions
1 Introduction
2 Preliminaries
3 Disagreement Options
3.1 Task Similarity: How to Select Relevant Priors?
3.2 Task Adaptation: How Should We Use the Prior Knowledge?
3.3 Prior Policy Acquisition
4 Experiments
4.1 3D MiniWorld
4.2 Photorealistic Simulator
5 Towards Real-World Task Adaptation
6 Related Work
7 Discussion
8 Conclusion
References
Deep Adaptive Multi-intention Inverse Reinforcement Learning
1 Introduction
2 Related Works
3 Problem Definition
4 Approach
4.1 First Solution with Stochastic Expectation Maximization
4.2 Second Solution with Monte Carlo Expectation Maximization
5 Experimental Results
5.1 Benchmarks
5.2 Models
5.3 Metric
5.4 Implementations Details
5.5 Results
6 Conclusions
References
Unsupervised Task Clustering for Multi-task Reinforcement Learning
1 Introduction
2 Related Work
3 Background and Notation
4 Clustered Multi-task Learning
4.1 Convergence Analysis
5 Experiments
5.1 Pendulum
5.2 Bipedal Walker
5.3 Atari
5.4 Ablations
6 Conclusion
References
Deep Model Compression via Two-Stage Deep Reinforcement Learning
1 Introduction
2 A Deep Reinforcement Learning Compression Framework
2.1 State
2.2 Action
2.3 Reward
2.4 The Proposed DRL Compression Structure
3 Pruning
3.1 Pruning from C Dimension: Channel Pruning
3.2 Pruning from H and W Dimensions: Variational Pruning
4 Quantization
5 Experiments
5.1 Settings
5.2 MNIST and CIFAR-10
5.3 ImageNet
5.4 Variational Pruning via Information Dropout
5.5 Single Layer Acceleration Performance
5.6 Time Complexity
6 Conclusion
References
Dropout's Dream Land: Generalization from Learned Simulators to Reality
1 Introduction
2 Related Works
2.1 Dropout
2.2 Domain Randomization
2.3 World Models
3 Dropout's Dream Land
3.1 Learning the Dream Environment
3.2 Interacting with Dropout's Dream Land
3.3 Training the Controller
4 Experiments
4.1 Comparison with Baselines
4.2 Inference Dropout and Dream2Real Generalization
4.3 When Should Dropout Masks Be Randomized During Controller Training?
4.4 Comparison to Standard Regularization Methods
4.5 Comparison to Explicit Ensemble Methods
5 Conclusion
References
Goal Modelling for Deep Reinforcement Learning Agents
1 Introduction
2 Background
3 Deep Reinforcement Learning with Goal Net
4 Experiments
4.1 Two Keys
4.2 3D Four Rooms with Subgoals
4.3 Kitchen Navigation and Interaction
5 Related Work
6 Discussion and Conclusion
References
Time Series, Streams, and Sequence Models
Deviation-Based Marked Temporal Point Process for Marker Prediction
1 Introduction
2 Related Work
3 Proposed Algorithm
3.1 Problem Definition
3.2 Preliminaries
3.3 Proposed Deviation-Based Marked Temporal Point Process
3.4 Implementation Details
4 Experiments and Protocols
5 Results and Analysis
6 Conclusion and Discussion
References
Deep Structural Point Process for Learning Temporal Interaction Networks
1 Introduction
2 Related Work
3 Background
3.1 Temporal Interaction Network
3.2 Temporal Point Process
4 Proposed Model
4.1 Overview
4.2 Embedding Layer
4.3 Topological Fusion Encoder
4.4 Attentive Shift Encoder
4.5 Model Training
4.6 Model Analysis
5 Experiments
5.1 Datasets
5.2 Experiment Setting
5.3 Item Prediction
5.4 Time Prediction
5.5 Discussion of Model Variants
6 Conclusion
References
Holistic Prediction for Public Transport Crowd Flows: A Spatio Dynamic Graph Network Approach
1 Introduction
2 Literature Review
2.1 In-Out Flow Prediction
2.2 OD Transit Flow Matrix Prediction
3 Problem Formulation
4 Methodology
4.1 Spatial Correlation
4.2 Temporal Correlation
4.3 Fusion
5 Experiments and Results
5.1 Experimental Settings
5.2 In-Out Flow Prediction
5.3 OD Transit Flow Prediction
5.4 Sensitivity Analysis
6 Conclusions
References
Reservoir Pattern Sampling in Data Streams
1 Introduction
2 Related Work
3 Preliminaries
4 Reservoir Algorithms for Pattern Sampling
4.1 Challenges and Key Ideas
4.2 Generic Algorithm: ResPat
4.3 Fast Damping Algorithms: ResPat no, ResPat win and ResPat exp
4.4 Theoretical Analysis
5 Experimental Evaluation
5.1 Global and Longitudinal Performance Study
5.2 Use Case: One-Pass Frequent Pattern Outlier Detection
6 Conclusion
References
Discovering Proper Neighbors to Improve Session-Based Recommendation
1 Introduction
2 Related Work
2.1 Collaborative Filtering-Based Methods
2.2 Graph Neural Networks-Based Methods
2.3 Memory Network-Based Methods
3 The Proposed Method: CoKnow
3.1 Problem Definition
3.2 Current Session Modeling with Item Tag (Cu-tag)
3.3 Neighbor Session Modeling with Item Tag (Ne-tag)
3.4 Prediction Layer
4 Experiments
4.1 Research Questions
4.2 Datasets
4.3 Baselines
4.4 Evaluation Metrics and Experimental Setup
4.5 Results and Analysis
5 Conclusion
References
Continuous-Time Markov-Switching GARCH Process with Robust State Path Identification and Volatility Estimation
1 Introduction
1.1 Motivation and Problem
1.2 Related Work
1.3 Our Contribution
2 COMS-GARCH Process
3 Inference for COMS-GARCH Process
3.1 Gibbs Sampler for Bayesian Inference on Model Parameters
3.2 Estimation for State Path and Volatility
3.3 Bernoulli Noise Injection
3.4 Theoretical Analysis on Inferential Benefits of Bernoulli NI
4 Experiments
4.1 Experiment Setting
4.2 Results
5 Discussion
References
Dynamic Heterogeneous Graph Embedding via Heterogeneous Hawkes Process
1 Introduction
2 Related Work
3 Preliminaries
4 The Proposed HPGE Model
4.1 Overview
4.2 Heterogeneous Conditional Intensity Modeling
4.3 Heterogeneous Evolved Attention Mechanism
4.4 Temporal Importance Sampling
4.5 Optimization Objective
5 Experiments
5.1 Experimental Settings
5.2 Effectiveness Analysis
5.3 Model Analysis
6 Conclusion
References
Explainable Online Deep Neural Network Selection Using Adaptive Saliency Maps for Time Series Forecasting
1 Introduction
2 Literature Review
3 Methodology
3.1 Preliminaries
3.2 Candidate CNN Architectures
3.3 Online Model Selection
4 Experiments
4.1 Experimental Setup
4.2 OS-PGSM Setup and Baselines
4.3 Results
4.4 Discussion and Future Work
5 Concluding Remarks
References
Change Detection in Multivariate Datastreams Controlling False Alarms
1 Introduction
2 Related Work
3 Problem Formulation
4 Proposed Solution
4.1 QuantTree Exponentially Weighted Moving Average
4.2 Computing Thresholds Controlling the ARL0
5 Datastream Monitoring by One-Shot Detectors
5.1 Monitoring Datastreams by Batch-Wise Detectors
5.2 Monitoring Datastreams by Element-Wise Detectors
6 Computational Complexity
7 Experiments
7.1 Considered Datasets
7.2 Figures of Merit
7.3 Results and Discussion
8 Conclusions
References
Approximation Algorithms for Confidence Bands for Time Series
1 Introduction
2 Preliminaries and Problem Definitions
3 Regularized Bands
3.1 Properties of Regularized Bands
3.2 Computing Regularized Band for a Single
3.3 Computing All Regularized Bands
4 Discovering Confidence Bands Minimizing bold0mu mumu ssbold0mu mumu 11
5 Discovering Confidence Bands Minimizing bold0mu mumu ssbold0mu mumu
6 Related Work
7 Experimental Evaluation
8 Concluding Remarks
References
A Mixed Noise and Constraint-Based Approach to Causal Inference in Time Series
1 Introduction
2 State of the Art
3 Causal Graph Discovery Between Two Time Series
3.1 Assumptions
3.2 Method
3.3 Complexity Analysis
4 Experiments
4.1 Simulated Data
4.2 Real Data
5 Conclusion
References
Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise
1 Introduction
2 Related Work
3 CHP Electrical Power Output Estimation
4 Self-Re-Labeling with Embedding Analysis (SREA)
5 Experimental Setup
6 Results and Discussion
6.1 Ablation Studies
7 Conclusion and Future Work
References
Multi-task Learning Curve Forecasting Across Hyperparameter Configurations and Datasets
1 Introduction
2 Problem Setting
3 Multi-LCNet: Multivariate Multi-step Forecasting with Meta-features
3.1 Optimizing Multi-LCNet
4 Experiments6
4.1 Datasets, Meta-Datasets and Evaluation Protocol
4.2 Baselines
4.3 Forecasting Results
4.4 Accelerating Hyperperameter Optimization
4.5 Acceleration Results
4.6 Ablation Study on the Meta-validation Set
5 Conclusion
References
Streaming Decision Trees for Lifelong Learning
1 Introduction
2 Related Works
3 Decision Trees and Lifelong Learning
3.1 Forgetting in Streaming Decision Trees
3.2 Overcoming Catastrophic Forgetting
4 Experimental Study
4.1 Data
4.2 Algorithms
4.3 Evaluation
4.4 Results
5 Summary
References
Transfer and Multi-task Learning
Unifying Domain Adaptation and Domain Generalization for Robust Prediction Across Minority Racial Groups
1 Introduction
2 Methodology
3 Experimental Setup
4 Multi-Source DA Across the Three Medical Systems
5 Domain Generalization to Unseen Medical System
6 Unseen Racial Group Across Medical Systems
7 Related Work
8 Conclusion
References
Deep Multi-task Augmented Feature Learning via Hierarchical Graph Neural Network
1 Introduction
2 Related Works
3 Hierarchical Graph Neural Network
3.1 Overview of the Architecture
3.2 The Model
3.3 Testing Process
3.4 Extension to Regression Tasks
3.5 Analysis
4 Experiments
4.1 Experimental Settings
4.2 Experimental Results
4.3 Ablation Study
4.4 Visualization
4.5 Sensitivity Analysis
5 Conclusion
References
Bridging Few-Shot Learning and Adaptation: New Challenges of Support-Query Shift
1 Introduction
2 The Support-Query Shift Problem
2.1 Statement
2.2 Positioning and Related Works
3 FewShiftBed: A Pytorch Testbed for FSQS
3.1 Datasets
3.2 Algorithms
3.3 Protocol
4 Transported Prototypes: A Baseline for FSQS
4.1 Overall Idea
4.2 Background
4.3 Method
5 Experiments
6 Conclusion
References
Source Hypothesis Transfer for Zero-Shot Domain Adaptation
1 Introduction
2 Related Work
3 Problem Setting and Background
3.1 Problem Setting
3.2 Ordinary Supervised Learning
3.3 ZSDA with Source Domain Data
4 Proposed Method
4.1 Model Collaboration
4.2 Hyperparameter Tuning
4.3 Implementation
5 Theoretical Analysis
5.1 Notations and Assumptions
5.2 Results
6 Experiments
6.1 Datasets
6.2 Setting
6.3 Evaluation Measure
6.4 Results
7 Conclusions
References
FedPHP: Federated Personalization with Inherited Private Models
1 Introduction
2 Related Works
3 Our Methods
3.1 Empirical Observation and Goal
3.2 Inherited Private Models
3.3 FedPHP
3.4 Discussion
3.5 Theoretical Analysis
4 Experiments
4.1 Scenes and Basic Settings
4.2 Experimental Results
4.3 Ablation Studies
5 Conclusion
References
Rumour Detection via Zero-Shot Cross-Lingual Transfer Learning
1 Introduction
2 Related Work
3 Methodology
3.1 Rumour Classifier
3.2 Cross-Lingual Transfer
4 Experiments and Results
4.1 Datasets
4.2 Experiment Setup
4.3 Results
4.4 Adaptive Pretraining and Layer Freezing
4.5 Self-training
4.6 Semi-supervised Learning
5 Discussion and Conclusions
References
Continual Learning with Dual Regularizations
1 Introduction
2 Related Work
3 Methodology
3.1 Problem Formulation
3.2 Proposed Approach
4 Experiments
4.1 Experimental Setting
4.2 Experimental Results
4.3 Ablation Study
5 Conclusions
References
EARLIN: Early Out-of-Distribution Detection for Resource-Efficient Collaborative Inference
1 Introduction
2 Related Work
3 Proposed OOD Detection Approach: EARLIN
4 Collaborative Inference Based on EARLIN
5 Experimental Evaluation of EARLIN
6 Prototype Implementation and Results
7 Conclusion and Future Works
References
Semi-supervised and Few-Shot Learning
LSMI-Sinkhorn: Semi-supervised Mutual Information Estimation with Optimal Transport
1 Introduction
2 Problem Formulation
3 Methodology
3.1 Least-Squares Mutual Information with Sinkhorn Algorithm
3.2 Optimization
3.3 Discussion
4 Related Work
5 Experiments
5.1 Setup
5.2 Convergence and Runtime
5.3 SMI Estimation
5.4 Deep Image Matching
5.5 Photo Album Summarization
6 Conclusion
References
Spatial Contrastive Learning for Few-Shot Classification
1 Introduction
2 Related Work
3 Preliminaries
3.1 Problem Definition
3.2 Transfer Learning Baseline
3.3 Analysis of the Learned Representations
4 Methodology
4.1 Contrastive Learning
4.2 Spatial Contrastive Learning
4.3 Pre-training Objective
4.4 Avoiding Excessive Disentanglement
5 Experiments
5.1 Experimental Details
5.2 Ablation Studies
5.3 Few-Shot Classification
5.4 Cross-Domain Few-Shot Classification
6 ProtoNet Experiments
7 Conclusion
References
Ensemble of Local Decision Trees for Anomaly Detection in Mixed Data
1 Introduction
2 Related Work
2.1 Methods for Numeric Data
2.2 Methods for Categorical Data
2.3 Ensemble Approaches
3 Our Proposal: An Ensemble of Local Decision Trees
3.1 Feating for Classification
3.2 Feating for Anomaly Detection
4 Experimental Results
4.1 Sensitivity of Parameters
5 Conclusions and Future Work
References
Learning Algorithms and Applications
Optimal Teaching Curricula with Compositional Simplicity Priors
1 Introduction
2 Notation and Background
3 Absolute Teaching Size and Complexity
4 Conditional Teaching Size
4.1 Conditional Teaching Size and Minimal Curriculum
4.2 Interposition and Non-monotonicity
5 Minimal Curriculum: Interposition Range and I-search
5.1 Interposition Range: I-sets
5.2 Teaching Size Upper Bounds: I-safe
5.3 Minimal Curriculum Algorithm: I-search
6 Conclusions and Future Work
References
FedDNA: Federated Learning with Decoupled Normalization-Layer Aggregation for Non-IID Data
1 Introduction
2 Related Work
3 Decoupled Federated Learning with CDCS
3.1 Optimization Objectives
3.2 FedDNA Mechanism
4 Adversarial Learning Algorithm
4.1 Adversarial Training
4.2 Inference with VAE
5 Performance Evaluation
5.1 Experimental Setup
5.2 Performance Analysis
6 Conclusion
References
The Curious Case of Convex Neural Networks
1 Introduction
2 Related Work
3 Input Output Convex Networks
3.1 Convexity as Self Regularizer
3.2 IOC-NN Decision Boundaries
3.3 Ensemble of IOC-NN
4 Experiments
4.1 Results
5 Conclusions
References
UCSL : A Machine Learning Expectation-Maximization Framework for Unsupervised Clustering Driven by Supervised Learning
1 Introduction
2 Related Works
3 UCSL: An Unsupervised Clustering Driven by Supervised Learning Framework
3.1 Mathematical Formulation
3.2 Expectation Step
3.3 Maximization Step
3.4 Supervised Predictions
3.5 Application
3.6 Pseudo-code
4 Results
5 Conclusion
References
Efficient and Less Centralized Federated Learning
1 Introduction
2 Preliminary and Related Work
2.1 Centralized Federated Learning: Federated Averaging
2.2 Multi-model Centralized Federated Learning
2.3 Decentralized Federated Learning
2.4 Efficient Pairwise Communication
3 Less Centralized Federated Learning
3.1 Proposed Framework: FedP2P
3.2 Communication Efficiency
3.3 Theoretical Insight
4 Experiments
4.1 Datasets
4.2 Implementation
4.3 Model Accuracy
4.4 Communication Efficiency
4.5 Stragglers Effect and Choice of L and Q
5 Conclusion
References
Topological Anomaly Detection in Dynamic Multilayer Blockchain Networks
1 Introduction
2 Related Work
3 The Mechanism of Persistent Homology
4 Persistence Methodology for Network Anomaly Detection
4.1 Theoretical Properties of the Stacked Persistence Diagram
5 Experiments on Blockchain Networks
5.1 Experimental Setup
5.2 Ethereum Token Networks
5.3 Ripple Currency Networks
6 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken