Black Box Model

Black box models, characterized by their opaque internal workings, pose challenges in understanding their decision-making processes, hindering trust and accountability. Current research focuses on improving interpretability through methods like generalized additive models (GAMs) and surrogate models, as well as addressing vulnerabilities to adversarial attacks and biases through techniques such as explanation-driven attacks and robust defense mechanisms. This work is crucial for building trust in AI systems across various applications, from medical diagnosis to autonomous driving, by enhancing transparency and mitigating potential risks associated with unpredictable model behavior.

Papers

November 11, 2022

REVEL Framework to measure Local Linear Explanations for black-box models: Deep Learning Image Classification case of study
Iván Sevillano-García, Julián Luengo-Martín, Francisco Herrera
Artificial Intelligence Image Classification Line by Line Explanation Study Feature Black Box Model Visual Explanation Local Explanation

November 3, 2022

Data-free Defense of Black Box Models Against Adversarial Attacks
Gaurav Kumar Nayak, Inder Khatri, Ruchit Rawal, Anirban Chakraborty
Adversarial Attack Black Box Model Based Attack

October 31, 2022

Scoring Black-Box Models for Adversarial Robustness
Jian Vora, Pranay Reddy Samala
Adversarial Attack Native Robustness Adversarial Robustness Black Box Model Adversarial Input Score Based

October 18, 2022

Explanations Based on Item Response Theory (eXirt): A Model-Specific Method to Explain Tree-Ensemble Model in Trust Perspective
José Ribeiro, Lucas Cardoso, Raíssa Silva, Vitor Cirilo, Níkolas Carneiro, Ronnie Alves
Line by Line Explanation XAI Method Black Box Model Model Based Item Response Theory XAI Research Tree Ensemble Model

October 10, 2022

Investigating the Failure Modes of the AUC metric and Exploring Alternatives for Evaluating Systems in Safety Critical Applications
Swaroop Mishra, Anjana Arunkumar, Chitta Baral
Black Box Model Large Pre Trained Model Area Under the ROC Curve Failure Mode Previously Generated Alternative Various Metric Safety Critical Application

October 4, 2022

Explanation-by-Example Based on Item Response Theory
Lucas F. F. Cardoso, José de S. Ribeiro, Vitor C. A. Santos, Raíssa L. Silva, Marcelle P. Mota, Ricardo B. C. Prudêncio, Ronnie C. O. Alves
Black Box Model xAI Community Item Response Theory Random Forest Model

October 3, 2022

Probabilistic Safeguard for Reinforcement Learning Using Safety Index Guided Gaussian Process Models
Weiye Zhao, Tairan He, Changliu Liu
Reinforcement Learning Black Box Model Safety Guarantee Model Learning Safe Controller Forward Invariance

September 30, 2022

September 28, 2022

September 19, 2022

A model-agnostic approach for generating Saliency Maps to explain inferred decisions of Deep Learning Models
Savvas Karatsiolis, Andreas Kamilaris
Deep Learning Model Artificial Intelligence Model Saliency Map Black Box Model Decision Relevant Information Model Agnostic

September 5, 2022

Visualization Of Class Activation Maps To Explain AI Classification Of Network Packet Captures
Igor Cherepanov, Alex Ulmer, Jonathan Geraldi Joewono, Jörn Kohlhammer
Deep Learning High Explainability Black Box Model Interactive Visualization Class Activation Network Analysis Network Data Deep Packet Inspection

September 3, 2022

Negative Selection Approach to support Formal Verification and Validation of BlackBox Models' Input Constraints
Abdul-Rauf Nuhu, Kishor Datta Gupta, Wendwosen Bellete Bedada, Mahmoud Nabil, Lydia Asrat Zeleke, Abdollah Homaifar, Edward Tunstel
Black Box Model External Validation Formal Verification Input Constraint

August 29, 2022

Interpreting Black-box Machine Learning Models for High Dimensional Datasets
Md. Rezaul Karim, Md. Shajalal, Alex Graß, Till Döhmen, Sisay Adugna Chala, Alexander Boden, Christian Beecks, Stefan Decker
Deep Neural Network High Dimensional Black Box Model High Dimension

August 26, 2022

PDD-SHAP: Fast Approximations for Shapley Values using Functional Decomposition
Arne Gevaert, Yvan Saeys
Shapley Value Black Box Model Fusion SHAP Efficient Approximation Function Decomposition Functional ANOVA

August 23, 2022

August 17, 2022

Quality Diversity Evolutionary Learning of Decision Trees
Andrea Ferigo, Leonardo Lucio Custode, Giovanni Iacca
Reinforcement Learning Inherent Interpretability Decision Tree Black Box Model Explainable Machine Learning

August 13, 2022

MaskBlock: Transferable Adversarial Examples with Bayes Approach
Mingyuan Fan, Cen Chen, Ximeng Liu, Wenzhong Guo
Adversarial Attack Adversarial Example Bayesian Inference Black Box Model Blackbox Attack

Black Box Model

Papers

REVEL Framework to measure Local Linear Explanations for black-box models: Deep Learning Image Classification case of study

Data-free Defense of Black Box Models Against Adversarial Attacks

Scoring Black-Box Models for Adversarial Robustness

Explanations Based on Item Response Theory (eXirt): A Model-Specific Method to Explain Tree-Ensemble Model in Trust Perspective

Investigating the Failure Modes of the AUC metric and Exploring Alternatives for Evaluating Systems in Safety Critical Applications

Explanation-by-Example Based on Item Response Theory

Probabilistic Safeguard for Reinforcement Learning Using Safety Index Guided Gaussian Process Models

Higher-order Neural Additive Models: An Interpretable Machine Learning Model with Feature Interactions

Online Multi-Agent Decentralized Byzantine-robust Gradient Estimation

MLink: Linking Black-Box Models from Multiple Domains for Collaborative Inference

Variance Tolerance Factors For Interpreting ALL Neural Networks

A model-agnostic approach for generating Saliency Maps to explain inferred decisions of Deep Learning Models

Visualization Of Class Activation Maps To Explain AI Classification Of Network Packet Captures

Negative Selection Approach to support Formal Verification and Validation of BlackBox Models' Input Constraints

Interpreting Black-box Machine Learning Models for High Dimensional Datasets

PDD-SHAP: Fast Approximations for Shapley Values using Functional Decomposition

SurvSHAP(t): Time-dependent explanations of machine learning survival models

Anomaly Attribution with Likelihood Compensation

Quality Diversity Evolutionary Learning of Decision Trees

MaskBlock: Transferable Adversarial Examples with Bayes Approach