Reasoning Web. Causality, Explanations and Declarative Knowledge: 18th International Summer School 2022, Berlin, Germany, September 27–30, 2022, Tutorial Lectures

Preface
Organization
Contents
Attribution-Scores and Causal Counterfactuals as Explanations in Artificial Intelligence
	1 Introduction
	2 The Role of Explanations in AI
	3 Some Classical Models of Explanation
		3.1 Consistency-Based Diagnosis
		3.2 Abduction
		3.3 Actual Causality and Responsibility
	4 Attribution Scores in Machine Learning
		4.1 The Generalized Resp Score
		4.2 The Shap Score
		4.3 Computation of the Shap Score
	5 Counterfactual Reasoning
	References
Logic-Based Explainability in Machine Learning
	1 Introduction
	2 Preliminaries
		2.1 Logic Foundations
		2.2 Classification Problems
		2.3 Non-formal Explanations
	3 Formal Explainability
		3.1 Abductive Explanations
		3.2 Contrastive Explanations
		3.3 Global Abductive Explanations and Counterexamples
		3.4 Duality Results
		3.5 Additional Notes
		3.6 A Timeline for Formal Explainability
	4 Computing Explanations
		4.1 Progress in Computing Explanations
		4.2 General Oracle-Based Approach
		4.3 Explaining Decision Lists
		4.4 From DLs to DTs and DSs
		4.5 Explaining Neural Networks
		4.6 Other Families of Classifiers
		4.7 An Alternative – Compilation-Based Approaches
	5 Tractable Explanations
		5.1 Decision Trees
		5.2 Monotonic Classifiers
		5.3 Other Families of Classifiers
	6 Explainability Queries
		6.1 Enumeration of Explanations
		6.2 Explanation Membership
		6.3 Additional Explainability Queries
	7 Probabilistic Explanations
		7.1 Problem Formulation
		7.2 Probabilistic Explanations for Decision Trees
		7.3 Additional Results
	8 Input Constraints and Distributions
	9 Formal Explanations with Surrogate Models
	10 Additional Topics and Extensions
	11 Future Research and Conclusions
		11.1 Research Directions
		11.2 Concluding Remarks
	References
Causal Inference in Data Analysis with Applications to Fairness and Explanations
	1 Introduction
	2 Pearl\'s Graphical Causal Model
		2.1 Graphical Models
		2.2 Structural, Graphical, and Probabilistic Causal Model
		2.3 Interventions, Do-Operators, and Counterfactuals
		2.4 Identification and Estimation
	3 Rubin\'s Potential Outcome Framework
		3.1 Observational Studies with Potential Outcome Framework and Ignorability
		3.2 Methods for Observational Causal Inference in Statistics
		3.3 Causal Inference on Networked and Relational Data
	4 Causal Algorithmic Fairness
		4.1 Associational Fairness
		4.2 Causal Fairness
	5 Causal Explanations for ML Models
	6 Conclusions
	References
Statistical Relational Extension of Answer Set Programming
	1 Introduction
	2 Review: Stable Model Semantics
	3 Language LPMLN
		3.1 Syntax of LPMLN
		3.2 Semantics of LPMLN (Reward-Based)
		3.3 Examples
		3.4 Alternative Formulation (Penalty-Based)
	4 Relation to Other Languages
		4.1 Relating LPMLN to ASP
		4.2 Relating LPMLN to MLNs
		4.3 Relating LPMLN to ProbLog
	5 Weight Learning
		5.1 Gradient Method for Learning Weights from a Complete Stable Model
		5.2 Sampling Method: MC-ASP
		5.3 Extensions
		5.4 Learning in the Presence of Noisy Data
		5.5 Learning from Incomplete Interpretations
	6 LPMLN System
	7 Multi-valued Probabilistic Programs
	8 Conclusion
	References
Vadalog: Overview, Extensions and Business Applications
	1 Introduction
	2 The Base
		2.1 A Map to Vadalog\'s Extensions and Business Applications
		2.2 A Gentle Introduction to Dependencies and Datalog
		2.3 From Datalog to Vadalog
	3 The Extensions
		3.1 Arithmetics and Aggregation
		3.2 Interfacing with the Real World
		3.3 Temporal Reasoning
		3.4 Machine Learning
	4 The Applications
		4.1 Corporate Governance
		4.2 Media Intelligence
		4.3 Supply Chains
		4.4 Collateral Eligibility
		4.5 Hostile Takeovers and Golden Powers
		4.6 Smart Anonymization
		4.7 Anti-Money Laundering (AML)
	References
Cross-Modal Knowledge Discovery, Inference, and Challenges
	1 Multimodal Knowledge Graph
		1.1 What is Multimodal Knowledge?
		1.2 Why Do We Need Multimodal Knowledge?
		1.3 Multimodal Knowledge Graph Construction
	2 Multimodal Knowledge Discovery
	3 Inference
	4 Future Research Opportunities and Challenges
		4.1 Knowledge Representation and Multimodal Learning
		4.2 NLP and Database
		4.3 IoT and Human-Computer Interaction
		4.4 Challenges
	References
Author Index