Eoin M. Kenny


eoin.kenny (at) jpmorgan (dot) com
Senior Associate AI Researcher
Trustworthy AI, J.P. Morgan

About

I am a Senior AI Researcher in the Trustworthy AI research group at J.P. Morgan in London. Previously, I was a postdoctoral researcher at MIT with Julie Shah, and completed my Ph.D. at University College Dublin with Mark Keane.

My research focuses on making AI systems understandable and useful to the people who work with them. I develop interpretability methods and evaluate whether they help people build accurate mental models, calibrate their reliance on AI, improve models, and make informed decisions. My work spans interpretable reinforcement learning, autonomous driving, scientific discovery, and contrastive explanations for algorithmic recourse.

Our recent research, published in Nature, shows that concept-based explanations can improve safety drivers' understanding of self-driving cars and their situational awareness. In collaboration with MIT and Motional, we deployed our technique on experimental autonomous vehicles on the Las Vegas Strip, with evaluations on closed tracks, public roads, and in online simulations. Learn more on the project website.

Recent News

Work Experience

Education

Selected Publications (from Google Scholar)

Explainable deep learning improves human mental models of self-driving cars

TL;DR: We show that concept-based explanations improve human mental models of self-driving cars, and that this translates into improve ability to predict the AV's future actions and general situational awareness.

Eoin M. Kenny, Akshay Dharmavaram, Sang Uk Lee, Tung Phan-Minh, Shreyas Rajesh, Yunqing Hu, Laura Major, Momchil S Tomov, Julie A. Shah

[Nature 2026]

Beyond Proxy Metrics: MLLM-Based Human Surrogate Evaluation for Explainable AI

TL;DR: We ask whether multimodal LLMs can stand in for human participants when evaluating XAI, and find that the rank ordering of explanation methods replicates across five purpose-grounded domains, though not uniformly.

Eoin M. Kenny, Salim I. Amoukou, Tom Bewley

Under review at NeurIPS 2026

CAMP: Coherent Alignment of Multimodal Prototypes for Explainable Complementary Learning

TL;DR: We build the first prototype-based interpretable-by-design framework for multimodal tasks where each modality carries complementary rather than redundant evidence, showing a sub-1M parameter head can rival AutoML systems 100x its size.

Alvaro Lopez Pellicer, Eoin M. Kenny, Simran Lamba, Shubham Sharma, Plamen P Angelov, Saumitra Mishra

[ICML 2026]

An Evaluation of Cost Functions for Algorithmic Recourse

TL;DR: We run the first large-scale benchmark of the cost functions used in algorithmic recourse, and show that Bradley-Terry models are best, but only when scaled up using LLMs as surrogate human labellers.

Eoin M. Kenny, Allan Anzagira, Tom Bewley, Freddy Lecue, Manuela Veloso

[AISTATS 2026]

A Hybrid Model that Combines Machine Learning and Mechanistic Models for Useful Grass Growth Prediction

TL;DR: We show mechanistic and machine learning models of grass growth fail in complementary ways, and that a screening model choosing between them instance-by-instance beats either alone, especially under out-of-distribution weather.

Eoin M. Kenny, Elodie Ruelle, Mark T. Keane, Laurence Shalloo

[Computers and Electronics in Agriculture 2024] * Code

The Utility of "Even if" Semifactual Explanation to Optimize Positive Outcomes

TL;DR: We show that semifactuals are more useful than conterfactuals when a user gets a positive outcome from an AI system.

Eoin M. Kenny, Weipeng Huang

[NeurIPS 2023]

Towards Interpretable Deep Reinforcement Learning with Human-Friendly Prototypes

TL;DR: We build the first inherently interpretable, general, well performaning, deep reinforcement learning algorithm.

Eoin M. Kenny, Mycal Tucker, and Julie A. Shah

[ICLR 2023] * Spotlight Presentation (top 25% of accepted papers)

On generating plausible counterfactual and semi-factual explanations for deep learning

TL;DR: We introduce the AI world to semi-factuals, and show a plausible way to generate them (and counterfactuals) using a framework called PIECE.

Eoin M. Kenny and Mark T. Keane

[AAAI 2022]

Bayesian Case-Exclusion and Explainable AI (XAI) for Sustainable Farming

TL;DR: We show how to accuractly predict grass growth and offer "good" explanations to Irish Dairy Farmers.

Eoin M Kenny, Elodie Ruelle, Anne Geoghegan, Mohammed Temraz, Mark T Keane

[IJCAI 2020] Sister Conference Best Paper Track * Best Paper Award at ICCBR 2019.

Explaining black-box classifiers using post-hoc explanations-by-example: The effect of explanations and error-rates in XAI user studies

TL;DR: We find that nearest neighbor exemplar-based explanation lead people to view classifiction errors as being “less incorrect”, moreover they do not effect trust.

Eoin M Kenny, Courtney Ford, Molly Quinn, Mark T Keane

[Artificial Intelligence 2021]

Media

National AI Awards Ireland

TL;DR: I won the best application of AI in a student project for my work in Explainable AI in Smart Agriculture.

[Agriland Press Release]

International Conference on Case-Based Reasoning

TL;DR: I won the best paper award at ICCBR 2019 which was themed on Explainable AI for my paper regarding Smart Agriculture.

Invited Talks (not exhaustive)

Academic Services