Below are my academic publications separated by topic.

Machine Learning

opt thumbnail

Adaptive versus Standard Descent Methods and Robustness Against Adversarial Examples
Marc Khoury
CoRR abs/1911.03784, 2019
arxiv / models

advtrain thumbnail

Adversarial Training with Voronoi Constraints
Marc Khoury, Dylan Hadfield-Menell
SafeML Workshop, International Conference on Learning Representations (ICLR), 2019

geomae thumbnail

On the Geometry of Adversarial Examples
Marc Khoury, Dylan Hadfield-Menell
CoRR abs/1811.00525, 2018

Computational Geometry

rcdt thumbnail

Restricted Constrained Delaunay Triangulations
Marc Khoury, Jonathan Shewchuk
Symposium on Computational Geometry (SoCG), 2021

approxtri thumbnail

Approximation Bounds for Interpolation and Normals on Triangulated Surfaces and Manifolds
Marc Khoury, Jonathan Shewchuk
CoRR abs/1911.03424, 2019

srp thumbnail

Supporting Ruled Polygons
Nicholas J. Cavanna, Marc Khoury, Donald R. Sheehy
Canadian Conference on Computational Geometry (CCCG), 2017

fixed point thumbnail

Fixed Points of the Restricted Delaunay Triangulation Operator
Marc Khoury, Jonathan Shewchuk
Symposium on Computational Geometry (SoCG), 2016

Computer Vision

lgf thumbnail

Learning Compact Geometric Features
Marc Khoury, Qian-Yi Zhou , Vladlen Koltun
International Conference on Computer Vision (ICCV), 2017
paper / supplement / code / data


dynamic graph drawing

Drawing Large Graphs by Low-Rank Stress Majorization
Marc Khoury, Yifan Hu, Shankar Krishnan, Carlos Scheidegger
Computer Graphics Forum
Proceedings of Eurographics Conference on Visualization (EuroVis), 2012
paper / code

fractal thumbnail

On the Fractal Dimension of Isosurfaces
Marc Khoury, Rephael Wenger
IEEE Transactions on Visualization and Computer Graphics (TVCG)
Proceedings of Visualization (Vis), 2010


Geometric Sampling Theory and Robust Machine Learning Algorithms
Jonathan Shewchuk
NSF Award CCF-1909235, $400,000

Based on my work on adversarial robustness above; I contributed a significant fraction of the project ideas and text.