Ruda Zhang

Ruda Zhang

Assistant Professor

University of Houston

Please check out the website—uq.uh.edu—for the most up-to-date information. It is the official website for my group, the Uncertainty Quantification Lab at the University of Houston.

My research combines machine learning (ML) and uncertainty quantification (UQ) for computational and data-enabled science and engineering (CDS&E).

I am interested in developing fast, reliable, and interpretable methods for data-driven engineering and uncertainty quantification, with applications in structural digital twins, advanced materials and manufacturing, and quantitative sustainability and resilience. See my research overview.

My previous work concerns the use of game theory and mechanism design in engineering and social systems, especially in urban transportation.

My curriculum vitae.

Interests

  • Uncertainty Quantification
  • Machine Learning
  • Data-driven Engineering
  • Computational Science & Engineering

Education

  • PhD, Civil Engineering, 2018

    University of Southern California

  • MA, Economics, 2018

    University of Southern California

  • BEng, Engineering Structure Analysis, 2012

    Peking University

Recent Papers

Gaussian Process Subspace Regression for Model Reduction

A fast and accurate method that approximates subspace-valued functions, with exampmle applications in projection-based parametric reduced order modeling.
Preprint PDF Code Project Slides Award
Gaussian Process Subspace Regression for Model Reduction

Newton Retraction as Approximate Geodesics on Submanifolds

Newton retraction is always stabler and faster than orthographic retraction, aka oblique projection.
PDF Project
Newton Retraction as Approximate Geodesics on Submanifolds

Normal-bundle Bootstrap

A method that generates new data which preserve the geometric structure of a given data set, with applications in inference and data augmentation.
PDF Code Project Slides Video DOI
Normal-bundle Bootstrap

Drivers Learn City-scale Dynamic Equilibrium

Simplistic, testable models of collective human behavior and dynamics can emerge from individual-level mechanisms that correctly capture incentives.
PDF Dataset Project DOI
Drivers Learn City-scale Dynamic Equilibrium

Demand, Supply and Performance of Street-Hail Taxi

Taxis out-perform TNCs such as Uber in high-demand locations, which suggests a taxi/TNC regulation change to reduce congestion and pollution.
PDF Dataset Project DOI
Demand, Supply and Performance of Street-Hail Taxi

Book Chapters

Environmental Economics and Uncertainty: Review and a Machine Learning Outlook

An integrated assessment of risks to coastal community, economy, and the ecosystem due to oil spills in the Gulf of Mexico, using a PLoM surrogate.
PDF Project DOI
Environmental Economics and Uncertainty: Review and a Machine Learning Outlook

Working Papers

Flow Surrogate of Computer Models

Manifold Surrogate of Computer Models

For computer models of complex physical and engineering systems, surrogates are often necessary to accelerate analyses that require large numbers of …

Accelerated Subspace-constrained Mean Shift

Subspace-constrained Mean Shift (SCMS) is an iterative algorithm for non-parametric ridge estimation. Although SCMS is computationally feasible, it is …
Project

Manifold Scaffold Sampling

When data points exhibit salient geometric structure, density estimation and generative modeling can be more efficient by exploiting the data …
Project

Kernel Density Estimation and Sampling on Riemmanian Submanifolds

For probability density estimation on Riemannian manifolds, many methods have been proposed, parametric or non-parametric. However, whether they can …
Project

Projects

*
All Research Software

manifold-GP

Learning manifold-valued functions using Gaussian process.

PLoM

Probabilistic Learning on Manifolds.

Scholar

Bibliographic analysis with Google Scholar.

Taxi

Analysis of the iconic yellow cabs in NYC.

Accessibility

Multi-modal analysis of Portland, OR.

Wiki

My personal knowledge system.

Contact