Josh Bone

Josh Bone

Data Scientist & Software Engineer

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

NFL Ratings - Introduction

3 minute read

Published:

======

Predicting NFL outcomes with an Elo rating system - Introduction

portfolio

publications

Electrical and Optical Modeling of Electrode Configuration for Optimal Dust Removal in Electrodynamic Screens (EDS)

Published in IEEE PVSC, 2019

This is a paper I wrote as an undergraduate at Boston University. It is about my modeling experiments which informed design choices in the Electrodynamic Screen thin film technology.

Recommended citation: Joshua Bone et. al, (2019). Electrical and Optical Modeling of Electrode Configuration for Optimal Dust Removal in Electrodynamic Screens (EDS). IEEE-PVSC 46 Conference

Geometry-Based Person Re-Identification in Fisheye Stereo

Published in 2021 17th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), 2021

This is another paper I published during my undergraduate years at Boston University. It describes a technique to leverage the fixed geometry of a scene in order to perform person re-identification with multiple overhead fisheye cameras.

Recommended citation: J. Bone, M. Cokbas, O. Tezcan, J. Konrad and P. Ishwar, "Geometry-Based Person Re-Identification in Fisheye Stereo," 2021 17th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), 2021, pp. 1-10, doi: 10.1109/AVSS52988.2021.9663745.

Analyzing Explainer Robustness via Probabilistic Lipschitzness of Prediction Functions

Published in AISTATS, 2024

This is a piece of theoretical work in explainable AI. It uses the concept of Lipschitzness from analysis to provide guarantees on the stability/robustness of explainer models.

Recommended citation: Z. Khan, D. Hill, A. Masoomi, J. Bone, and J. Dy. “Analyzing Explainer Robustness via Probabilistic Lipschitzness of Prediction Functions”. In: Proceedings of The 27th International Conference on Artificial Intelligence and Statistics 238 (May 2024), pp. 1378–1386.

Axiomatic Explainer Globalness via Optimal Transport

Published in AISTATS, 2025

This is the research that I presented during my PhD qualifying exams.

Recommended citation: J. Bone, D. Hill, A. Masoomi, M. Torop, and J. Dy. “Axiomatic Explainer Globalness via Optimal Transport”. In: Proceedings of The 28th International Conference on Artificial Intelligence and Statistics. Ed. by Y. Li, S. Mandt, S. Agrawal, and E. Khan. Vol. 258. Proceedings of Machine Learning Research. PMLR, Mar. 2025, pp. 1351–1359. url: <https://proceedings.mlr.press/v258/hill25a.html>. https://proceedings.mlr.press/v258/hill25a.html

talks