I am a mathematician/machine learning researcher at Pacific Northwest National Lab and a Affiliate Assistant Professor in the University of Washington Mathematics Department. My research interests include:
- Math for AI: Geometric, topological and algebraic approaches to deep learning,
- Real-world robustness for AI: Methods that produce safer, more secure, more robust, and more explainable deep learning models.
- Science of deep learning: Building an empirically grounded understanding of deep learning models, including their failure modes,
- AI for math: Machine learning methods to accelerate mathematics research.
During the academic year, I organize the 'Pacific Northwest Seminar on Topology, Algebra, and Geometry in Data Science' at the University of Washington. We are hybrid, please join us!
I am also one of three founding organizers of the Topology, Algebra, and Geometry in Data Science series of Workshops, Conferences, and Journal Special Editions.
News and blogposts featuring my work:
- Some thoughts on the role of mathematics in AI research
- Our recent paper on the intrinsic dimension of hidden activations in Stable Diffusion was mentioned in the Gradient
- Machine Learning Accelerates Development of Advanced Manufacturing Techniques
- Artificial Intelligence Tools for Advanced Manufacturing Processes
- Using AI to Explore Possibilities in Advanced Manufacturing
We recently released a set of datasets for applying ML to open and classic problems in algebraic combinatorics, 'Benchmarks in Algebraic Combinatorics'!
Publications + Preprints
Preprints
Most of my work is available at preprints on arXiv.
2024
Machines and Mathematical Mutations: Using GNNs to Characterize Quiver Mutation Classes
Jesse He, Helen Jenne, Herman Chau, Davis Brown, Mark Raugas, Sara Billey, Henry Kvinge
The 4th Workshop on Mathematical Reasoning and AI at NeurIPS'24
Machine Learning meets Algebraic Combinatorics: A Suite of Datasets to Accelerate AI for Mathematics Research
Herman Chau, Helen Jenne, Davis Brown, Jesse He, Mark Raugas, Sara Billey, Henry Kvinge
The 4th Workshop on Mathematical Reasoning and AI at NeurIPS'24
What Makes a Machine Learning Task a Good Candidate for an Equivariant Network?
Scott Mahan, Davis Brown, Timothy Doster, Henry Kvinge
ICML 2024 Workshop on Geometry-grounded Representation Learning and Generative Modeling
Model editing for distribution shifts in uranium oxide morphological analysis
Davis Brown, Cody Nizinski, Madelyn Shapiro, Cory Fallon, Tianzhixi Yin, Henry Kvinge, Jonathan Tu
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
Generalist Multimodal AI: A Review of Architectures, Challenges and Opportunities
Sai Munikoti, Ian Stewart, Sameera Horawalavithana, Henry Kvinge, Tegan Emerson, Sandra Thompson, Karl Pazdernik
Wild Comparisons: A Study of how Representation Similarity Changes when Input Data is Drawn from a Shifted Distribution
Davis Brown, Madelyn Shapiro, Alyson Bittner, Jackson Warley, Henry Kvinge
ICLR 2024 Workshop on Representational Alignment
2023
Internal Representations of Vision Models Through the Lens of Frames on Data Manifolds
Henry Kvinge, Grayson Jorgenson, Davis Brown, Charles Godfrey, Tegan Emerson
NeurIPS 2023 Workshop on Symmetry and Geometry in Neural Representations
Haldane Bundles: A Dataset for Learning to Predict the Chern Number of Line Bundles on the Torus
Cody Tipton, Elizabeth Coda, Davis Brown, Alyson Bittner, Jung Lee, Grayson Jorgenson, Tegan Emerson, Henry Kvinge
AI for Accelerated Materials Design-NeurIPS 2023 Workshop
Can We Count on Deep Learning: Exploring and Characterizing Combinatorial Structures Using Machine Learning
Helen Jenne, Herman Chau, Davis Brown, Jackson Warley, Timothy Doster, Henry Kvinge
The 3rd Workshop on Mathematical Reasoning and AI at NeurIPS'23
Understanding the Inner Workings of Language Models Through Representation Dissimilarity
Davis Brown, Charles Godfrey, Nick Konz, Jonathan Tu, Henry Kvinge
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Attributing Learned Concepts in Neural Networks to Training Data
Nick Konz, Charles Godfrey, Madelyn Shapiro, Jonathan Tu, Henry Kvinge, Davis Brown
ATTRIB Workshop at NeurIPS 2023
SCITUNE: Aligning Large Language Models with Scientific Multimodal Instructions
Sameera Horawalavithana, Sai Munikoti, Ian Stewart, Henry Kvinge, Henry Kvinge
ATTRIB Workshop at NeurIPS 2023
ColMix - A Simple Data Augmentation Framework to Improve Object Detector Performance and Robustness in Aerial Images
Cuong Ly, Grayson Jorgenson, Dan Rosa de Jesus, Henry Kvinge, Adam Attarian, Yijing Watkins
How many dimensions are required to find an adversarial example?
Charles Godfrey, Henry Kvinge, Elise Bishoff, Myles Mckay, Davis Brown, Tim Doster, Eleanor Byler
To appear at the The 3rd Workshop of Adversarial Machine Learning on Computer Vision: Art of Robustness
Fast computation of permutation equivariant layers with the partition algebra
Charles Godfrey, Michael Rawson, Davis Brown, Henry Kvinge
To appear at the ICLR 2023 Workshop on Physics for Machine Learning
Robustness of edited neural networks
Davis Brown, Charles Godfrey, Cody Nizinski, Jonathan Tu, Henry Kvinge
To appear at the ICLR 2023 Workshop on Mathematical and Empirical Understanding of Foundation Models
Exploring the Representation Manifolds of Stable Diffusion Through the Lens of Intrinsic Dimension
Henry Kvinge, Davis Brown, Charles Godfrey,
To appear at the ICLR 2023 Workshop on Mathematical and Empirical Understanding of Foundation Models
2022
Do neural networks trained with topological features learn different internal representations?
Sarah McGuire, Shane Jackson, Tegan Emerson, Henry Kvinge
NeurIPS Workshop on Symmetry and Geometry in Neural Representations, 122-136, 2023
Parameters, Properties, and Process: Conditional Neural Generation of Realistic SEM Imagery Toward ML-Assisted Advanced Manufacturing
Scott Howland, Lara Kassab, Keerti Kappagantula, Henry Kvinge, Tegan Emerson
Integrating Materials and Manufacturing Innovation, 1-10, 2023
Convolutional networks inherit frequency sensitivity from image statistics
Charles Godfrey, Elise Bishoff, Myles Mckay, Davis Brown, Grayson Jorgenson, Henry Kvinge, Eleanor Byler
In What Ways Are Deep Neural Networks Invariant and How Should We Measure This?
Henry Kvinge, Tegan H. Emerson, Grayson Jorgenson, Scott Vasquez, Timothy Doster, Jesse D. Lew
NeurIPS 2022
On the Symmetries of Deep Learning Models and their Internal Representations
Charles Godfrey, Davis Brown, Tegan Emerson, Henry Kvinge
NeurIPS 2022
Making Corgis Important for Honeycomb Classification:
Adversarial Attacks on Concept-based Explainability Tools
Davis Brown, Henry J Kvinge
2022 ICML Workshop on New Frontiers in Adversarial Machine Learning
Random Filters for Enriching the Discriminatory Power of Topological Representations
Tegan Emerson, Grayson Jorgenson, Henry Kvinge, Colin Olson
2022 ICLR Workshop on Geometric and Topological Representation Learning
TopTemp: Parsing Precipitate Structure from Temper Topology
Tegan Emerson, Lara Kassab, Scott Howland, Henry Kvinge, Keerti Sahithi, Kappagantula
2022 ICLR Workshop on Geometric and Topological Representation Learning
Fiber Bundle Morphisms as a Framework for Modeling Many-to-Many Maps
Elizabeth Coda, Nico Courts, Colby Wight, Loc Truong, WoongJo Choi, Charles Godfrey, Tegan Emerson, Keerti Kappagantula, Henry Kvinge
2022 ICLR Workshop on Geometric and Topological Representation Learning
Bundle Networks
Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps
Nico Courts, Henry Kvinge
International Conference on Learning Representations. 2021
Differential Property Prediction
A Machine Learning Approach to Experimental Design in Advanced Manufacturing
Loc Truong, WoongJo Choi, Colby Wight, Lizzy Coda, Tegan Emerson, Keerti Kappagantula, Henry Kvinge
2021 AAAI Workshop on AI for Design and Manufacturing (ADAM)
2021
Hypergraph models of biological networks to identify genes critical to pathogenic viral response
Song Feng, Emily Heath, Brett Jefferson, Cliff Joslyn, Henry Kvinge, Hugh D Mitchell, Brenda Praggastis, Amie J Eisfeld, Amy C Sims, Larissa B Thackray, Shufang Fan, Kevin B Walters, Peter J Halfmann, Danielle Westhoff-Smith, Qing Tan, Vineet D Menachery, Timothy P Sheahan, Adam S Cockrell, Jacob F Kocher, Kelly G Stratton, Natalie C Heller, Lisa M Bramer, Michael S Diamond, Ralph S Baric, Katrina M Waters, Yoshihiro Kawaoka, Jason E McDermott, Emilie Purvine
BMC Bioinformatics
Sheaves as a Framework for Understanding and Interpreting Model Fit
Henry Kvinge, Brett Jefferson, Cliff Joslyn, Emilie Purvine
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops
Multi-dimensional scaling on groups
Mark Blumstein, Henry Kvinge
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops
A Topological-Framework to Improve Analysis of Machine Learning Model Performance
Henry Kvinge, Colby Wight, Sarah Akers, Scott Howland, Woongjo Choi, Xiaolong Ma, Luke Gosink, Elizabeth Jurrus, Keerti Kappagantula, Tegan H Emerson
ICML 2021 Workshop on Uncertainty and Robustness in Deep Learning
Fuzzy Simplicial Networks
A Topology-Inspired Model to Improve Task Generalization in Few-shot Learning
Henry Kvinge, Zachary New, Nico Courts, Jung H Lee, Lauren A Phillips, Courtney D Corley, Aaron Tuor, Andrew Avila, Nathan O Hodas
AAAI Workshop on Meta-Learning and MetaDL Challenge
One Representation to Rule Them All
Identifying Out-of-Support Examples in Few-shot Learning with Generic Representations
Henry Kvinge, Scott Howland, Nico Courts, Lauren A. Phillips, John Buckheit, Zachary New, Elliott Skomski, Jung H. Lee, Sandeep Tiwari, Jessica Hibler, Courtney D. Corley, Nathan O. Hodas
Rotating spiders and reflecting dogs
A class conditional approach to learning data augmentation distributions
Scott Mahan, Henry J Kvinge, Tim Doster
Adaptive Transfer Learning: a simple but effective transfer learning
Jung H Lee, Henry J Kvinge, Scott Howland, Zachary New, John Buckheit, Lauren A Phillips, Elliott Skomski, Jessica Hibler, Courtney D Corley, Nathan O Hodas
DNA: Dynamic Network Augmentation
Scott Mahan, Tim Doster, Henry Kvinge
2020
Prototypical Region Proposal Networks for Few-Shot Localization and Classification
Elliott Skomski, Aaron Tuor, Andrew Avila, Lauren Phillips, Zachary New, Henry Kvinge, Courtney D. Corley, Nathan Hodas
2020 NeurIPS Workshop on Meta-Learning
The center of the twisted Heisenberg category, factorial Schur Q-functions, and transition functions on the Schur graph
Henry Kvinge, Can Ozan Oğuz, Michael Reeks
Journal of Algebraic Combinatorics
Dimensionality Reduction
Sofya Chepushtanova, Elin Farnell, Eric Kehoe, Michael Kirby, Henry Kvinge
Chapter 7 from Data Science for Mathematicians
Tayor and Francis
LWIR compressive sensing hyperspectral sensor for chemical plume imaging
Julia R Dupuis, John P Dixon, Elizabeth Schundler, S Chase Buchanan, JD Rameau, David Mansur, Henry Kvinge, Elin Farnell, Chris Peterson, Michael Kirby
SPIE - Defense + Commercial Sensing
More chemical detection through less sampling: amplifying chemical signals in hyperspectral data cubes through compressive sensing
Henry Kvinge, Elin Farnell, Julia R Dupuis, Michael Kirby, Chris Peterson, Elizabeth C Schundler
SPIE - Defense + Commercial Sensing
A data-driven approach to sampling matrix selection for compressive sensing
Elin Farnell, Henry Kvinge, John P Dixon, Julia R Dupuis, Michael Kirby, Chris Peterson, Elizabeth C Schundler, Christian W Smith
SPIE - Defense + Commercial Sensing
Total variation vs L1 regularization: a comparison of compressive sensing optimization methods for chemical detection
Elin Farnell, Henry Kvinge, Julia R Dupuis, Michael Kirby, Chris Peterson, Elizabeth C Schundler
SPIE - Defense + Commercial Sensing
Mathematical methods for visualization and anomaly detection in telemetry datasets
Manuchehr Aminian, Helene Andrews-Polymenis, Jyotsana Gupta, Michael Kirby, Henry Kvinge, Xiaofeng Ma, Patrick Rosse, Kristin Scoggin, David Threadgill
Interface Focus, The Royal Society
Mathematical methods for visualization and anomaly detection in telemetry datasets
Lucius Bynum, Timothy Doster, Tegan H Emerson, Henry Kvinge
IGARSS 2020-2020 IEEE International Geoscience and Remote Sensing Symposium,
2019
Rare geometries: revealing rare categories via dimension-driven statistics
Henry Kvinge, Elin Farnell, Jingya Li, Yujia Chen
2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA),
A Walk Through Spectral Bands: Using Virtual Reality to Better Visualize Hyperspectral Data
Henry Kvinge, Henry Kvinge, Michael Kirby, Chris Peterson, Chad Eitel, Tod Clapp
International Workshop on Self-Organizing Maps,
Khovanov’s Heisenberg category, moments in free probability, and shifted symmetric functions
Henry Kvinge, Anthony M Licata, Stuart Mitchell
Algebraic Combinatorics,
2018
Monitoring the shape of weather, soundscapes, and dynamical systems: a new statistic for dimension-driven data analysis on large datasets
Henry Kvinge, Elin Farnell, Michael Kirby, Chris Peterson
2018 IEEE International Conference on Big Data (Big Data)2018 IEEE International Conference on Big Data (Big Data),
A Combinatorial Categorification of the Tensor Product of the Kirillov-Reshetikhin Crystal B1,1 and a Fundamental Crystal
Henry Kvinge, Monica Vazirani
Algebras and Representation Theory,
Too many secants: a hierarchical approach to secant-based dimensionality reduction on large data sets
Henry Kvinge, Elin Farnell, Michael Kirby, Chris Peterson
2018 IEEE High Performance extreme Computing Conference (HPEC),
A GPU-Oriented Algorithm Design for Secant-Based Dimensionality Reduction
Henry Kvinge, Elin Farnell, Michael Kirby, Chris Peterson
2018 17th International Symposium on Parallel and Distributed Computing (ISPDC),
Endmember Extraction on the Grassmannian
Elin Farnell, Henry Kvinge, Michael Kirby, Chris Peterson
2018 IEEE Data Science Workshop (DSW 2018),
A Frobenius-Schreier-Sims Algorithm to tensor decompose algebras
Ian Holm Kessler, Henry Kvinge, James B Wilson
Coherent systems of probability measures on graphs for representations of free Frobenius towers
Henry Kvinge
Any opinions expressed here are personal, and do not necessarily reflect those of my employer.