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 16 of 16 for “"Post-hoc Explanation"”.
-
Exploring neural network interpretability in visual understanding
… NN interpretability in diverse visual tasks from post-hoc explanation and intrinsic interpretability perspectives. Convolutional neural networks (CNNs) have outperformed humans in image classification. However, the logic of network decisions remains a puzzle. As such, we propose concept-harmonized …
-
POWER AND COHERENCE MATTERS: DOES PERCEIVED GROUP ORGANIZATION INCREASE ATHEIST PREJUDICE?
… decreases prejudice towards atheists. One post-hoc explanation provided for this difference is that since atheists are easily concealable and unorganized as a group, their greater prevalence may not be perceived as a threat. In the present thesis, I 1) attempted to replicate the existing …
-
The relationship between extent of aftereffect and speed of a rotated spiral
… or the function of speed on the aftereffect. A post hoc explanation was offered to explain the effect of speed on spiral aftermovement by an extension of the statistical summation theory. The following prediction was based on pilot data with the intent to determine if speed was a stable general …
-
Transformation-invariance properties of Grad-CAM in Convolutional Neural Networks
… Class Activation Mapping (Grad-CAM) is a post-hoc explanation technique that generates visual heatmaps to highlight the class-relevant regions in an input image. However, in real-world scenarios, input images often undergo transformations such as rotation, zoom, and shifts along the …
-
Towards Reliable and Trustworthy Deep Learning through Explainability and Interpretability
… of deep learning models. First, we develop new post-hoc explanation methods that improve feature attribution and concept-based explanations. These methods provide faithful decision cues by modeling meaningful feature interactions and extracting faithful coherent concepts, enabling more reliable …
-
Unifying Explainability and Machine Learning: Towards Trustworthy and Plausible Explanations without Post-hoc Manipulations
… and enhance model interpretability, ensuring explanations are both faithful (accurate to the model’s reasoning) and plausible (intelligible to human users). A key challenge in XAI is the unreliability of post-hoc explanation methods. To address this, we propose (1) quantitative evaluation …
-
Improving the interpretability of machine learning approaches with user-generated data
… of the model overcomes the limitations of common post-hoc explanation techniques. Furthermore, since the resulting explanations are faithful to the model reasoning, the results can also be used for model debugging and hyperparameter tuning.
-
Model Explainability: From Post-hoc to Self-explainable
… through three complementary approaches: improved post-hoc explanation methods, internal mechanism understanding, and inherently interpretable architectures. First, we tackle the challenge of explaining video understanding models by proposing Frequency-based Extremal Perturbation (F-EP), which …
-
Towards Explainable Time Series: An Exploration Of Post-hoc Local Explanations From A Multi-disciplinary Perspective
… the Explainable AI landscape, post-hoc local explanations, which help users understand why a black-box returned a particular output for a particular input, have become the most widely applied tool. Despite their success, post-hoc explanations are still in their infancy and …
-
Physical Side-Channel Vulnerability Assessment of Implementations of Cryptographic Algorithms
… features from the NN models using gradient-based post hoc explanation methods to the ones obtained by traditional points of interest (PoI) selection methods. Third, we performed a power side-channel vulnerability assessment of a parallel hardware implementation of one of the finalists of the NIST …
-
The impact of cultural values on email acceptance : evidence from the PRC
… typology and to remedy the inadequacy of post hoc explanation in conventional IS cultural studies. Instead of using the scores of national culture proposed by Hofstede’s cultural study in 1980, this study measures cultural values in terms of personal traits to reflect the changes in and …
-
Hipiess: A Framework for Artificial Intelligence and Machine Learning in Clinical Informatics
… (XAI) methods such as feature attribution and post-hoc explanation techniques offer partial insight into model behavior, they often fail to align with clinical reasoning, are not actionable, and are rarely integrated into clinical workflows. Finally, deployment systems such as clinical decision …
-
Causal and system-theoretic approaches to interpretable machine learning
… addresses these challenges by exploring both post-hoc explanation methods and augmentation based strategies for increasing interpretability. We begin by examining existing explainable AI (XAI) methods, which aim to attribute a model’s output to its input features. Focusing on the widely used …
-
Algorithmic Decision-Making Facilities: Perception and Design of Explainable AI-based Decision Support Systems
… of AI-based DSSs. This yields a plethora of explanation techniques, creating an XAI jungle from which end-users must choose. In turn, these techniques are preliminarily engineered by developers for developers without ensuring an actual end-user fit. Thus, it renders unknown how an end-user’s …
-
New and Advanced Explainable Artificial Intelligence Approaches for Managing Project Risks
… among risk factors, often producing fragmented explanations that overlook the structural dependencies underlying overall project vulnerability. In PRM, this limitation is particularly critical as risks are inherently interdependent, and disturbances in one aspect can propagate across multiple …
-
PROBABILISTIC MODEL CHECKING WITH MARKOV MODELS SEMANTICS: NEW DEVELOPMENTS AND APPLICATIONS
… our focus to a specific XAI framework, i.e., post-hoc explanation methods based on surrogate models. More specifically, we pro- pose a logic and related model-checking procedures for specifying and checking relevant reliability properties of XAI-explanations for opaque ML systems provided via …