Trustworthy Machine Learning

Trustworthy machine learning (TML) aims to build and deploy machine learning models that are reliable, robust, and fair, addressing concerns about accuracy, bias, and explainability. Current research focuses on improving model robustness against adversarial attacks and distributional shifts, enhancing interpretability through techniques like Shapley value analysis and knowledge distillation, and developing methods for quantifying and managing uncertainty. These advancements are crucial for increasing the adoption of AI in high-stakes applications like healthcare and finance, where trust in model outputs is paramount.

Papers

February 17, 2023

Function Composition in Trustworthy Machine Learning: Implementation Choices, Insights, and Questions
Manish Nagireddy, Moninder Singh, Samuel C. Hoffman, Evaline Ju, Karthikeyan Natesan Ramamurthy, Kush R. Varshney
DCU Insight AQ Yes No Question Practical Implementation Trustworthy Machine Learning Comprehensive Trustworthiness Bias Mitigation Algorithm Function Composition

February 6, 2023

Trust, but Verify: Using Self-Supervised Probing to Improve Trustworthiness
Ailin Deng, Shen Li, Miao Xiong, Zhirui Chen, Bryan Hooi
Appropriate Trust Trustworthy Machine Learning Comprehensive Trustworthiness Self Supervised Probing

January 31, 2023

Learning Against Distributional Uncertainty: On the Trade-off Between Robustness and Specificity
Shixiong Wang, Haowei Wang, Jean Honorio
LeArning Abstract Native Robustness Regularization Model Robust Optimization Trustworthy Machine Learning Distributional Uncertainty

December 18, 2022

A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness
Tiantian Feng, Rajat Hebbar, Nicholas Mehlman, Xuan Shi, Aditya Kommineni, and Shrikanth Narayanan
Machine Learning Procedural Fairness Privacy Policy Human SAFETY Machine Learning System Trustworthy Machine Learning

October 10, 2022

FEAMOE: Fair, Explainable and Adaptive Mixture of Experts
Shubham Sharma, Jette Henderson, Joydeep Ghosh
Expert Knowledge Interpretable Model Fairness Metric Fairness Constraint Trustworthy Machine Learning Adaptive Mixture

September 30, 2022

Empowering the trustworthiness of ML-based critical systems through engineering activities
Juliette Mattioli, Agnes Delaborde, Souhaiel Khalfaoui, Freddy Lecue, Henri Sohier, Frederic Jurie
Machine Learning Machine Learning System Position Engineering Trustworthy Machine Learning Data Engineering

September 14, 2022

Data Privacy and Trustworthy Machine Learning
Martin Strobel, Reza Shokri
Native Robustness Machine Learning Model Privacy Risk Data Privacy Trustworthy Machine Learning

July 27, 2022

Towards Clear Expectations for Uncertainty Estimation
Victor Bouvier, Simona Maggio, Alexandre Abraham, Léo Dreyfus-Schmidt
Uncertainty Quantification Uncertainty Estimation Trustworthy Machine Learning Expectation Alignment Uncertainty Score

July 7, 2022

Calibrate to Interpret
Gregory Scafarto, Nicolas Posocco, Antoine Bonnefoy
High Explainability Calibration Performance Confidence Calibration Trustworthy Machine Learning Interpretation Method

May 2, 2022

A Survey on Uncertainty Toolkits for Deep Learning
Maximilian Pintz, Joachim Sicking, Maximilian Poretschkin, Maram Akila
Deep Learning Timely Survey Uncertainty Quantification Uncertainty Estimation Tensorflow Probability Trustworthy Machine Learning Better Uncertainty Estimation Model Reliability

March 14, 2022

Closing the Loop: A Framework for Trustworthy Machine Learning in Power Systems
Jochen Stiasny, Samuel Chevalier, Rahul Nellikkath, Brynjar Sævarsson, Spyros Chatzivasileiadis
Machine Learning New Framework Power System Trustworthy Machine Learning Decarbonization Study Wind Farm Power Transformer

February 25, 2022

Bayesian autoencoders with uncertainty quantification: Towards trustworthy anomaly detection
Bang Xiang Yong, Alexandra Brintrup
Uncertainty Quantification Unsupervised Anomaly Detection Trustworthy Machine Learning Deep Autoencoders Total Uncertainty

February 15, 2022

Trustworthy Anomaly Detection: A Survey
Shuhan Yuan, Xintao Wu
Timely Survey Anomaly Detection Trustworthy Machine Learning Anomaly Detection Model

January 26, 2022

TrustAL: Trustworthy Active Learning using Knowledge Distillation
Beong-woo Kwak, Youngwook Kim, Yu Jin Kim, Seung-won Hwang, Jinyoung Yeo
Knowledge Distillation Active Learning Anti Forgetting Data Labeling Trustworthy Machine Learning

January 4, 2022

Trusting Machine Learning Results from Medical Procedures in the Operating Room
Ali El-Merhi, Helena Odenstedt Hergés, Linda Block, Mikael Elam, Richard Vithal, Jaquette Liljencrantz, Miroslaw Staron
Machine Learning Trustworthy Machine Learning Physiological Data Physiological Time Series Data Medical Procedure Operating Room Carotid Atherosclerosis