Explainability Problem

The explainability problem in machine learning focuses on understanding how and why complex models, particularly deep learning models like transformers and autoencoders, arrive at their predictions. Current research emphasizes developing methods to generate interpretable explanations, often using techniques like Layer-wise Relevance Propagation (LRP), attention maps, and concept-based attribution, while also analyzing the inherent complexities of explanation generation through parameterized complexity analysis. Addressing this problem is crucial for building trust in AI systems, improving model debugging and refinement, and ensuring responsible deployment across various applications, including medicine and natural language processing.

Papers

October 9, 2024

Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap
Georgia Channing, Juil Sock, Ronald Clark, Philip Torr, Christian Schroeder de Witt
Explainability Method Audio Deepfake Detection Audio Deepfake Audio DeepFakes Explainability Problem

October 7, 2024

Explanation sensitivity to the randomness of large language models: the case of journalistic text classification
Jeremie Bogaert, Marie-Catherine de Marneffe, Antonin Descampe, Louis Escouflaire, Cedrick Fairon, Francois-Xavier Standaert
Large Language Model High Explainability Case Relevance Inherent Randomness Composite Variable Construction Explainability Problem Stable Explanation Headline Classification

July 22, 2024

Explaining Decisions in ML Models: a Parameterized Complexity Analysis
Sebastian Ordyniak, Giacomo Paesani, Mateusz Rychlicki, Stefan Szeider
Machine Learning Model Explainable AI Line by Line Explanation Decision Relevant Information Parameterized Complexity Explainability Problem

February 28, 2024

Editing Factual Knowledge and Explanatory Ability of Medical Large Language Models
Derong Xu, Ziheng Zhang, Zhihong Zhu, Zhenxi Lin, Qidong Liu, Xian Wu, Tong Xu, Wanyu Wang, Yuyang Ye, Xiangyu Zhao, Enhong Chen, Yefeng Zheng
Large Language Model Model Editing Edited Knowledge Explainability Problem

November 1, 2023

Notion of Explainable Artificial Intelligence -- An Empirical Investigation from A Users Perspective
AKM Bahalul Haque, A. K. M. Najmul Islam, Patrick Mikalef
Explainable AI Empirical Study Explainable Artificial Intelligence Recommendation System User Perception Explainability Problem Human Centered Explainable AI

July 12, 2023

Learning from Exemplary Explanations
Misgina Tsighe Hagos, Kathleen M. Curran, Brian Mac Namee
LeArning Abstract Model Explanation Quality Explanation Interactive Machine Learning Explainability Problem Medical Image Classification Task Explanation Based

June 8, 2023

Sound Explanation for Trustworthy Machine Learning
Kai Jia, Pasapol Saowakon, Limor Appelbaum, Martin Rinard
Line by Line Explanation Trustworthy Machine Learning Attribution Based Explainability Problem Cancer Prediction

May 17, 2023

Interpretable Graph Networks Formulate Universal Algebra Conjectures
Francesco Giannini, Stefano Fioravanti, Oguzhan Keskin, Alisia Maria Lupidi, Lucie Charlotte Magister, Pietro Lio, Pietro Barbiero
Inherent Interpretability Topological Representation Explainability Problem Universal Algebra Interpretable Graph Neural Network

May 11, 2023

COCKATIEL: COntinuous Concept ranKed ATtribution with Interpretable ELements for explaining neural net classifiers on NLP tasks
Fanny Jourdan, Agustin Picard, Thomas Fel, Laurent Risser, Jean Michel Loubes, Nicholas Asher
Inherent Interpretability Transformer Model NLP Task Neural Network Classifier Explainability Problem Interpretability Benefit Concept Attribution NLP Classification Task

March 21, 2023

Unlocking Layer-wise Relevance Propagation for Autoencoders
Kenyu Kobayashi, Renata Khasanova, Arno Schneuwly, Felix Schmidt, Matteo Casserini
Supervised Autoencoder Layer Wise Relevance Propagation Explainability Problem Deep Taylor Decomposition

March 18, 2023

ExplainFix: Explainable Spatially Fixed Deep Networks
Alex Gaudio, Christos Faloutsos, Asim Smailagic, Pedro Costa, Aurelio Campilho
Deep Convolutional Neural Network Deep Network Spatial Convolution Explainability Problem Explainer Model

March 16, 2023

Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations
Changqing Qiu, Fusheng Jin, Yining Zhang
Fine Grained NCD Method Capability Evolution Explainability Problem Explanation Faithfulness Noisy Explanation

January 24, 2023

Explainable Deep Reinforcement Learning: State of the Art and Challenges
George A. Vouros
Reinforcement Learning Technical Challenge High Explainability Internal State Art Specific Information Explainable Reinforcement Learning Explainability Problem Explainable Deep Reinforcement Learning

September 12, 2022

Explaining Predictions from Machine Learning Models: Algorithms, Users, and Pedagogy
Ana Lucic
Machine Learning Machine Learning Model Practical Algorithm Human Prediction High Explainability User Base Model Explanation Pedagogical Alignment Explainability Problem

September 8, 2022

Towards explainable evaluation of language models on the semantic similarity of visual concepts
Maria Lymperaiou, George Manoliadis, Orfeas Menis Mastromichalakis, Edmund G. Dervakos, Giorgos Stamou
Language Model Pre Trained Language Model Evaluation Metric Semantic Similarity Visual Concept NLP Research Linguistic Representation Explainability Problem

September 3, 2022

Explainability via Short Formulas: the Case of Propositional Logic with Implementation
Reijo Jaakkola, Tomi Janhunen, Antti Kuusisto, Masood Feyzbakhsh Rankooh, Miikka Vilander
High Explainability Case Relevance Practical Implementation Propositional Logic Mathematical Formula Explainability Problem Polynomial Bound

Explainability Problem

Papers

Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap

Explanation sensitivity to the randomness of large language models: the case of journalistic text classification

Explaining Decisions in ML Models: a Parameterized Complexity Analysis

Editing Factual Knowledge and Explanatory Ability of Medical Large Language Models

Notion of Explainable Artificial Intelligence -- An Empirical Investigation from A Users Perspective

Learning from Exemplary Explanations

Sound Explanation for Trustworthy Machine Learning

Interpretable Graph Networks Formulate Universal Algebra Conjectures

COCKATIEL: COntinuous Concept ranKed ATtribution with Interpretable ELements for explaining neural net classifiers on NLP tasks

Unlocking Layer-wise Relevance Propagation for Autoencoders

ExplainFix: Explainable Spatially Fixed Deep Networks

Empowering CAM-Based Methods with Capability to Generate Fine-Grained and High-Faithfulness Explanations

Explainable Deep Reinforcement Learning: State of the Art and Challenges

Explaining Predictions from Machine Learning Models: Algorithms, Users, and Pedagogy

Towards explainable evaluation of language models on the semantic similarity of visual concepts

Explainability via Short Formulas: the Case of Propositional Logic with Implementation