Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 35 for “"Counterfactual Explanations"”.
-
Generation and evaluation of semantic counterfactual explanations
… of complex models. This dissertation focuses on counterfactual explanations, which determine the minimal input modifications required to alter an AI model’s prediction. A domain-agnostic, black-box framework for counterfactual generation is introduced, applicable across multiple data modalities, …
-
On evaluating counterfactual explanations for Machine Learning Models
… Learning – ML) μέσω αντιπαραθετικών εξηγήσεων (counterfactual explanations – CF), οι οποίες περιλαμβάνουν την πραγματοποίηση μικρών μεταβολών στα δεδομένα εισόδου προκειμένου να διερευνηθούν εναλλακτικά αποτελέσματα. Αναγνωρίζοντας τη διαρκώς αυξανόμενη σημασία της αξιολόγησης της απόδοσης …
-
Responsible decision-making in trustworthy human-AI interaction: linking counterfactual explanations and regret
… the somatic marker hypothesis driven by the counterfactual component of the anticipated regret could serve as a recommendation norm for preventing unspecified errors before they occur. Chapter 1 introduces the dissertation topic and justifies the motivations for conducting it. Chapter 2 …
-
Causality concepts in machine learning: heterogeneous treatment effect estimation with machine learning & model interpretation with counterfactual and semi-factual explanations
… concepts to interpret machine learning models: counterfactual explanations and semi-factual explanations. Counterfactual explanations describe minimal changes in a few features required for changing a prediction, while semi-factual explanations describe maximal changes in a few features required …
-
Multi-objective approaches towards trustworthy machine learning
… data changes over time, and methods to provide explanations for model decisions need to operate efficiently and in real-time. In this thesis, we address these challenges by developing frameworks that can account for more than one characteristic of responsible artificial intelligence. First, we …
-
What does explainable AI explain?
… aspects beyond predictive performance through explanations. However, the field is still in its infancy, as fundamental questions such as “What are explanations?”, “What constitutes a good explanation?”, or “How relate explanation and understanding?” remain open. In this dissertation, I combine …
-
Supporting Human Performance with Post-hoc Explanations in Automated Decision Assistance
… promote appropriate reliance behaviors. Post-hoc explanations have emerged as a strategy to support appropriate reliance on automated decision aids based on machine learning. However, existing methods to generate post-hoc explanations often fail to demonstrate systematic effectiveness in aiding …
-
Personalized Algorithmic Recourse: Towards a human-centric approach for algorithmic contestability
… individuals to challenge, appeal, or seek explanations for AI-driven decisions. Indeed, while legal frameworks, such as the EU AI Act, conceptualize contestability, they fail to describe technically or scientifically actionable methodologies for effective implementation. In machine …
-
PRINCIPLED ANALYSIS OF MODEL EXPLANATIONS
… While user studies are ideal for evaluating explanations in particular contexts, they are costly and impractical at scale. This thesis adopts a complementary approach by focusing on high-level properties and axioms that explanations should satisfy. First, we employ axiomatic characterization …
-
Explaining Expert Search and Team Formation Systems with ExES
… systems. Our system leverages saliency and counterfactual methods from the field of explainable artificial intelligence (XAI). ExES enables users to understand why individuals were or were not included in the query results and what individuals could do, in terms of perturbing skills or …
-
Counterfactuals for Interpretable Machine Learning: Model Reasoning from “What” to “How”
… in reasoning decisions and understanding "how". Counterfactual Explanation, serving as one of the emerging types of ML interpretations, has raised the attention from both researchers and practitioners in recent years. Counterfactual explanation is essentially a series of hypothetical data …
-
Algorithmic approaches for efficiency and explainability in learning problems
… to noisy and incomplete data. (2) For global counterfactual explanations, the problem is formulated as a multi-objective optimization balancing interpretability, cost, and effectiveness. Its computational hardness is formally established, and efficient clustering-based algorithms are …
-
Novel mathematical optimization Models for Group counterfactual Analysis
… Such modified solutions, the so-called counterfactual explanations, are addressed in this thesis. This dissertation is devoted to extending the study of counterfactual analysis across different dimensions. Specially, we delve into group analysis, where, instead of considering a single …
Page 1 of 2