Research / Publications
Scientific machine learning · Uncertainty quantification · Sensitivity analysis
Founder of SVD Lab. Previously, I was a Senior Research Scientist at Bosch and Apple, following postdoctoral research at MIT. I received my PhD from Tsinghua University.
My research focuses on scientific machine learning and uncertainty quantification, with an emphasis on learning physical dynamics, differentiable simulation and sensitivity analysis.
28 publications
No publications match this search. Try a different method or clear the research-area filter.
Combustion and Flame, 251, 112732 · 2023 Journal article
Treats kinetic parameter optimization as neural ODE training to calibrate hydrocarbon-fuel models against experimental targets.
Combustion and Flame, 240, 111992 · 2022 Journal article
Learns biomass pyrolysis kinetics from thermogravimetric measurements with interpretable chemical reaction neural networks.
Fuel, 324, 124560 · 2022 Journal article
Uses differentiable simulation and stochastic gradient descent to optimize mechanistic and hybrid chemical kinetic models.
arXiv · 2021 Preprint
Introduces differentiable combustion modeling in Julia for gradient-based sensitivity analysis, calibration, uncertainty quantification and model discovery.
The Journal of Physical Chemistry A, 125(4), 1082-1092 · 2021 Journal article
Infers interpretable reaction pathways and kinetic parameters from concentration time series using a neural architecture based on mass-action and Arrhenius laws.
ASME ICEF 2021 · 2021 Conference paper
Applies stochastic gradient descent to calibrate lumped HyChem fuel models against ignition-delay measurements across temperature regimes.
ICLR 2021 Workshop on Deep Learning for Simulation (SimDL) · 2021 Workshop paper
Uses adjoint sensitivity and differentiable programming to infer cell-network dynamics from simulated perturbation-response data.
arXiv · 2021 Preprint
Represents time-stepping chemical kinetics with a residual neural network and multi-step training, with ignition-delay gradients for model refinement.
ASME IMECE 2021 · 2021 Conference paper
Infers unknown inlet composition and flow fluctuations from sparse combustor-temperature measurements using neural differential equations.
Chaos: An Interdisciplinary Journal of Nonlinear Science, 31(9), 093122 · 2021 Journal article
Studies stiff neural ODE training using scale-aware architectures, output and loss scaling, and stabilized gradient calculations.
The Journal of Physical Chemistry A, 125(36), 8098-8106 · 2021 Journal article
Examines PINN failure on stiff chemical kinetics and uses quasi-steady-state reduction to make the studied systems tractable.
Proceedings of the Combustion Institute, 39(4), 5229-5238 · 2023 Journal article
Uses a neural-network surrogate to identify multidimensional kinetic subspaces for uncertainty propagation across nonpremixed flamelet profiles.
Proceedings of the Combustion Institute, 39(4), 5299-5308 · 2023 Journal article
Examines effective independent-data counts and target selection when using shock-tube species time histories for Bayesian kinetic parameter estimation.
Combustion and Flame, 227, 135-146 · 2021 Journal article
Combines active subspaces with an intermediate reduction state to separate uncertainty changes caused by parameter truncation and reaction coupling.
Combustion and Flame, 220, 16-22 · 2020 Journal article
Investigates how kinetic sensitivity directions vary through premixed flame structures and their connection to laminar flame speed and extinction.
Proceedings of the Combustion Institute, 37(1), 807-815 · 2019 Journal article
Relates the sensitivity directions of temperature, species and ignition delay, enabling efficient evaluation of ignition-delay sensitivities.
Proceedings of the Combustion Institute, 37(2), 2175-2182 · 2019 Journal article
Uses active subspaces and response surfaces to propagate kinetic uncertainty to the liftoff height of a turbulent Cabra hydrogen jet flame.
arXiv · 2019 Preprint
Builds response surfaces in gradient-based active subspaces to propagate uncertain neural-network inputs at reduced computational cost.
Combustion and Flame, 190, 146-157 · 2018 Journal article
Combines single-output active subspaces into a shared low-dimensional representation for propagating kinetic uncertainty to multiple quantities of interest.
10th U.S. National Combustion Meeting · 2017 Conference paper
Explores a low-rank surrogate subspace for propagating kinetic uncertainty, using neural response surfaces and genetic optimization.
Energy and AI, 7, 100128 · 2022 Journal article
Reviews machine-learning approaches to combustion modeling, experiments and computational prediction.
Nature Energy, 5(9), 666-673 · 2020 Journal article
Combines pandemic scenarios, mobility data and machine learning to project US motor-gasoline demand during COVID-19.
Proceedings of the Combustion Institute, 36(1), 315-322 · 2017 Journal article
Measures ignition delays of a fully blended low-octane light naphtha fuel.
Combustion and Flame, 183, 30-38 · 2017 Journal article
Studies rapid-compression-machine measurement of a reaction rate using dimethyl carbonate decomposition as a case study.
Proceedings of the Combustion Institute, 36(1), 343-353 · 2017 Journal article
Investigates crossover temperature and the lower turnover state of negative-temperature-coefficient ignition behavior.
Combustion and Flame, 167, 14-23 · 2016 Journal article
Combines experiments and simulations to investigate negative-temperature-coefficient behavior in first-stage ignition delay.
Combustion and Flame, 164, 294-302 · 2016 Journal article
Studies the kinetic mechanisms controlling the upper turnover states in the negative-temperature-coefficient regime.
Combustion and Flame, 162(10), 3541-3553 · 2015 Journal article
Measures intermediate species during iso-butanol autoignition to examine the underlying oxidation pathways.
Research files
Each paper has a permanent page with its citation, research summary and available full text. Download the complete metadata or browse the repository for research with Codex and other tools.