Explainability Technique

Explainability techniques aim to make the decision-making processes of complex machine learning models more transparent and understandable. Current research focuses on developing and comparing various methods for interpreting model outputs, including those based on feature importance, perturbation analysis, and natural language rationalization, across diverse model architectures such as transformers and graph neural networks. This work is crucial for building trust in AI systems, improving model debugging and performance, and enabling responsible deployment in high-stakes applications like healthcare and finance. A key challenge lies in evaluating the effectiveness and consistency of these techniques, particularly across different datasets and model types.

Papers

November 2, 2024

A Mechanistic Explanatory Strategy for XAI
Marcin Rabiza
Explainable AI xAI Community Mechanistic Model XAI Research Opaque Machine Learning Explainability Technique

October 28, 2024

Explainability in AI Based Applications: A Framework for Comparing Different Techniques
Arne Grobrugge, Nidhi Mishra, Johannes Jakubik, Gerhard Satzger
Artificial Intelligence New Framework High Explainability Explainability Method Interpretable AI Explainability Technique

October 20, 2024

Explainability of Point Cloud Neural Networks Using SMILE: Statistical Model-Agnostic Interpretability with Local Explanations
Seyed Mohammad Ahmadi, Koorosh Aslansefat, Ruben Valcarce-Dineiro, Joshua Barnfather
Point Cloud Inherent Interpretability Explainable AI High Explainability Explainability Method Model Agnostic Local Explanation Happy Image Explainability Technique

October 17, 2024

Interpreting Inflammation Prediction Model via Tag-based Cohort Explanation
Fanyu Meng, Jules Larke, Xin Liu, Zhaodan Kong, Xin Chen, Danielle Lemay, Ilias Tagkopoulos
Line by Line Explanation High Explainability Explainability Technique Expert Delineated Cohort

May 31, 2024

Model Interpretation and Explainability: Towards Creating Transparency in Prediction Models
Donald Kridel, Jacob Dineen, Daniel Dolk, David Castillo
Explainable AI High Explainability Transparency Index Feature Importance Model Explanation Prediction Model Explainability Technique Model Interpretation

May 17, 2024

Data Science Principles for Interpretable and Explainable AI
Kris Sankaran
Inherent Interpretability Explainable AI Data Science Interpretable Way Interpretability Method Explainability Technique Interpretable Algorithm

March 27, 2024

Enhancing Manufacturing Quality Prediction Models through the Integration of Explainability Methods
Dennis Gross, Helge Spieker, Arnaud Gotlieb, Ricardo Knoblauch
Machine Learning Predictive Model Explainability Method Explainability Technique Milling Process

November 20, 2023

On the Relationship Between Interpretability and Explainability in Machine Learning
Benjamin Leblanc, Pascal Germain
Machine Learning Inherent Interpretability High Explainability Human Relationship Explainability Technique Explainability Tool

November 8, 2023

Explained anomaly detection in text reviews: Can subjective scenarios be correctly evaluated?
David Novoa-Paradela, Oscar Fontenla-Romero, Bertha Guijarro-Berdiñas
Anomaly Detection Multi Scenario Text Review Explainability Technique Explainable Anomaly Detection

October 3, 2023

GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking
Mert Kosan, Samidha Verma, Burouj Armgaan, Khushbu Pahwa, Ambuj Singh, Sourav Medya, Sayan Ranu
Counterfactual Explanation Gene Level GNN Task Utility Benchmark Study Explainability Method GNN Explanation Explainability Technique Perturbation Based

September 2, 2023

Explainability for Large Language Models: A Survey
Haiyan Zhao, Hanjie Chen, Fan Yang, Ninghao Liu, Huiqi Deng, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, Mengnan Du
Large Language Model Natural Language Processing Timely Survey High Explainability Transformer Based Language Model Explanation Method Explainability Technique

May 25, 2023

Explainability Techniques for Chemical Language Models
Stefan Hödl, William Robinson, Yoram Bachrach, Wilhelm Huck, Tal Kachman
Self Attention Chemical Data Explainability Technique Interpretable Prediction Chemical Language Model Molecular String Representation

May 8, 2023

XAI in Computational Linguistics: Understanding Political Leanings in the Slovenian Parliament
Bojan Evkoski, Senja Pollak
Machine Learning Model xAI Community Computational Linguistics Left Leaning Explainability Technique Parliamentary Corpus Political Leaning

February 27, 2023

CrystalBox: Future-Based Explanations for Input-Driven Deep RL Systems
Sagar Patel, Sangeetha Abdu Jyothi, Nina Narodytska
Deep Reinforcement Learning Deep RL Post Hoc Quality Explanation Explainability Technique Future Based Explanation

January 21, 2023

Rationalization for Explainable NLP: A Survey
Sai Gurrapu, Ajay Kulkarni, Lifu Huang, Ismini Lourentzou, Laura Freeman, Feras A. Batarseh
Natural Language Processing Timely Survey Natural Language Explanation Model Explanation Rational Agent Explainability Technique Folded Rationalization

July 26, 2022

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics
Yiqiao Li, Jianlong Zhou, Sunny Verma, Fang Chen
Graph Neural Network Timely Survey Comprehensive Taxonomy Evaluation Metric Explainability Technique Explainable Graph

March 30, 2022

Explainability in Process Outcome Prediction: Guidelines to Obtain Interpretable and Faithful Models
Alexander Stevens, Johannes De Smedt
High Explainability State Aware Guideline Explanation Model Explainability Technique Faithful Model

February 21, 2022

Explainability in Machine Learning: a Pedagogical Perspective
Andreas Bueff, Ioannis Papantonis, Auste Simkute, Vaishak Belle
Machine Learning High Explainability Explainability Technique Pedagogical Alignment Transparent Machine Learning

February 20, 2022

Trying to Outrun Causality with Machine Learning: Limitations of Model Explainability Techniques for Identifying Predictive Variables
Matthew J. Vowels
Machine Learning Causal Structure Fundamental Limitation Causal Pattern Underlying Causal Explainability Technique Causal Objective Function

February 16, 2022

XAI in the context of Predictive Process Monitoring: Too much to Reveal
Ghada Elkhawaga, Mervat Abuelkheir, Manfred Reichert
Context Information xAI Community Process Mining eXplainable Artificial Intelligence Predictive Process Monitoring Explainability Technique Efficient Prediction

Explainability Technique

Papers

A Mechanistic Explanatory Strategy for XAI

Explainability in AI Based Applications: A Framework for Comparing Different Techniques

Explainability of Point Cloud Neural Networks Using SMILE: Statistical Model-Agnostic Interpretability with Local Explanations

Interpreting Inflammation Prediction Model via Tag-based Cohort Explanation

Model Interpretation and Explainability: Towards Creating Transparency in Prediction Models

Data Science Principles for Interpretable and Explainable AI

Enhancing Manufacturing Quality Prediction Models through the Integration of Explainability Methods

On the Relationship Between Interpretability and Explainability in Machine Learning

Explained anomaly detection in text reviews: Can subjective scenarios be correctly evaluated?

GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking

Explainability for Large Language Models: A Survey

Explainability Techniques for Chemical Language Models

XAI in Computational Linguistics: Understanding Political Leanings in the Slovenian Parliament

CrystalBox: Future-Based Explanations for Input-Driven Deep RL Systems

Rationalization for Explainable NLP: A Survey

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics

Explainability in Process Outcome Prediction: Guidelines to Obtain Interpretable and Faithful Models

Explainability in Machine Learning: a Pedagogical Perspective

Trying to Outrun Causality with Machine Learning: Limitations of Model Explainability Techniques for Identifying Predictive Variables

XAI in the context of Predictive Process Monitoring: Too much to Reveal