Jarvis Computational Mathematics Lab

Department of Mathematics, Brigham Young University

Principal Investigator: Tyler J. Jarvis, Professor of Mathematics

The Jarvis Computational Mathematics Lab, directed by Tyler J. Jarvis, develops rigorous mathematical and computational methods for real-world problems. Our work is currently organized around four research directions: numerical algebraic geometry, redistricting analysis, physiological signal processing, and generalized aliasing.

Research Directions

Numerical Algebraic Geometry

Developing fast, reliable numerical rootfinding methods for systems of polynomial and analytic equations, and the open-source Yroots software that implements them.

Redistricting Analysis

Applying mathematical and statistical methods to detect gerrymandering and evaluate the fairness of electoral district maps.

Physiological Signals

Mathematical modeling and signal-processing methods for periodic and pulsatile physiological signals, with applications to noninvasive health monitoring.

Generalized Aliasing

A new theoretical paradigm for understanding model complexity and generalization, explaining phenomena such as double descent and informing model design.

Research Direction

Numerical Algebraic Geometry

The lab develops numerical methods for finding all roots of systems of polynomial and analytic equations, with an emphasis on reliability, speed, and scalability to high dimensions. This work is implemented in Yroots, our open-source numerical rootfinding tool.

Research Direction

Redistricting Analysis

The lab studies the mathematics of political redistricting, developing quantitative tools to identify gerrymandering and to assess the fairness and compactness of proposed district maps.

Research Direction

Physiological Signals

The lab develops mathematical models and signal-processing algorithms for analyzing periodic and pulsatile physiological signals (e.g., cardiac and respiratory waveforms, spectrometer signals), with applications to noninvasive health monitoring.

Research Direction

Generalized Aliasing

The lab is developing generalized aliasing as a new theoretical paradigm for understanding how model complexity relates to predictive performance. This framework explains phenomena such as double descent that are not well captured by the classical bias–variance decomposition, and it can inform model design and experimental design before data collection.

People

Current Lab Members

Faculty Collaborators

Funding

Research by the PI has been supported by federal grants for three decades, including:

See the PI’s curriculum vitae for a complete list of funding, publications, and awards.

Contact

For questions about the lab, contact the principal investigator, Tyler J. Jarvis, Professor of Mathematics, Brigham Young University: jarvis@mathematics.byu.edu. Official profile: BYU Faculty Directory.