• July 2025: I gave two talks at USNCCM18 on learning history- and microstructure-dependent constitutive laws: (i) in MS 233, I presented ongoing work on Bayesian optimal experimental design [slides], and (ii) in the special session on SciML, I presented an operator learning approach from microscale simulations with a focus on theoretical results [slides]. Open the slides in Adobe Acrobat for animations.

• March 2025: I presented virtually at the Frontiers in Scientific Machine Learning Seminar Series, hosted by the University of Michigan, Ann Arbor, on March 28 from 12–1 PM EST. The talk was on derivative-informed neural surrogates and their applications in accelerating Bayesian inversion. [slides]

• March 2025: I presented at SIAM CSE25 in MS281 on Friday at 9:50 am @ 206 on learning history- and microstructure-dependent constitutive laws using operator learning and optimal experimental design [slides][preprint]. Open the slides in Adobe Acrobat for animations.

• December 2024: The special issue on Scientific Machine Learning is published in Foundations of Data Science. I am one of the guest editors of this special issue. [link]

• October 2024: I spoke at SIAM MDS24 in MS5 on Monday at 10:20 am, Room 218, on "Derivative-Informed Operator Learning and Its Application to Accelerating Large-Scale Bayesian Inverse Problems." [slides][preprint]

• November 2024: Our preprint titled "LazyDINO: Fast, scalable, and efficiently amortized Bayesian inversion via structure-exploiting and surrogate-driven measure transport" is now available on arXiv. This is joint work with Joshua Chen, Michael Brennan, Tom O'Leary-Roseberry, Youssef Marzouk, and Omar Ghattas. [link]

• October 2024: DC and I organized a minisymposium and a miniposterium at SIAM MDS24. The mini-symposium (MS5, Monday 9:00--10:40 am @ 218) focuses on Scientific Machine Learning for Inference and Control of High-Dimensional Systems. The miniposterium (Monday, 4:30 PM - 6:00 PM @ Grand Ballroom A-B) focused on Derivative-Informed Operator Learning. 

• September 2024: I gave a talk at MORe 2024 on derivative-informed operator learning and its application to accelerating large-scale Bayesian inverse problems. [slides][preprint]

• March 2024: Our preprint titled "Derivative-informed neural operator acceleration of geometric MCMC for infinite-dimensional Bayesian inverse problems" is now available on arXiv. This is joint work with Tom O'Leary-Roseberry and Omar Ghattas. [link]

• February 2024: Our manuscript, "Bayesian model calibration for block copolymer self-assembly: Likelihood-free inference and expected information gain via measure transport", is now published in Journal of Computational Physics. Congratulations to Ricardo, Josh, Fengyi, Omar, Youssef, and Dr. Oden! [link]

2023

• Dec. 2023: I gave an Oden Institute Seminar talk on Dec. 12. I presented ongoing work on efficient geometric MCMC enabled by derivative-informed neural operators. The paper is available on arXiv. [preprint]

• Nov. 2023: I gave a talk at the SIAM TX-LA regional meeting in MS29. I presented ongoing work on efficient geometric MCMC for PDE-constrained Bayesian inversion enabled by neural operators with parametric derivative training. The paper is available on arXiv. [preprint]

• Nov. 2023: The members of OPTIMUS and M2dt (Bassel, DC, and I) organized an MS at SIAM TX-LA titled "Mathematical and Computational Foundations of Predictive Digital Twins" (MS 33). We invited speakers from the research areas of model reduction, optimal control, and scientific machine learning to participate in our MS. The MS also included excellent talks on oceanic, biological, and robotic systems.

• Aug. 2023: Our manuscript, "Bayesian model calibration for diblock copolymer thin film self-assembly using power spectrum of microscopy data and machine learning surrogate", is now published in Computer Methods in Applied Mechanics and Engineering. Congratulations to Keyi, Peng, Dr. Oden, and Omar! This manuscript is dedicated to Prof. Tom Hughes for his lifelong contributions to Computational Sciences and Engineering. [link]

• Aug. 2023: I gave an invited talk at ICIAM 2023 Tokyo. The talk was in the "Scientific Machine Learning for Inverse Problems" session. I presented works on error estimation and correction for neural operator acceleration of PDE-constrained Bayesian inverse problems. [slides][preprint] [journal]

• Aug. 2023: I gave a lunch seminar at Caltech CMS on Aug. 2. I presented works on deriving self-consistent field theory for diblock copolymers and Bayesian parameter inference from microscopy data of diblock copolymer thin films. [slides] [thesis] [preprint]

• Jul. 2023: The paper "A nonlocal theory of heat transfer and micro-phase separation of nanostructured copolymers" was published in the International Journal of Heat and Mass Transfer. Congratulations to Pratyush, Danial, and Kira! [link]

• Jun. 2023: A preprint paper titled "Bayesian model calibration for diblock copolymer thin film self-assembly using power spectrum of microscopy data and machine learning surrogate" is now available. This paper is based on a chapter of my Ph.D. dissertation. [link]

• Jul. 2023: I gave a contributed talk at USNCCM17. The talk is in MS403: "Uncertainty Quantification for Learning and Data-Driven Predictive Modeling of Complex Systems". I will present works on Bayesian predictive modeling of diblock copolymer thin film self-assembly. [slides] [preprint] 

• Apr. 2023: A preprint paper titled "A nonlocal theory of heat transfer and micro-phase separation of nanostructured copolymers" is now available. [link] 

• Mar. 2023: I gave an invited talk at APS March Meeting [website] in Las Vegas, in the focus session on UQ for soft matter physics. I presented work on Bayesian predictive modeling for block copolymer self-assembly using triangular transport maps. [slides] [preprint]

• Mar. 2023: I gave an invited talk at SIAM CSE 2023 in Amsterdam. The talk is in the session on additive manufacturing. The session is featured on the "Industry At-a-Glance" flyer under the Design & Manufacturing section. I presented work on UQ using image characterizations of polymer thin films, based on my dissertation research.

• Mar. 2023: Two papers are accepted for publication in the Journal of Computational Physics (on the same day)! Congrats to DC, Tom, Prashant, Peng, Dr. Oden, and Omar!

• Feb. 2023: My dissertation won the 2023 Oden Institute Outstanding Dissertation Award! [thesis][news]

2022

• November 2022: I successfully defended my dissertation! [slides]

• November 2022: I gave an invited talk at the 5th Annual Meeting of the SIAM TX-LA Section, in MS: Recent Advances in Learning [website]. I presented results in our recent paper [preprint] on understanding and improving the reliability of neural operators as surrogates of parametric PDEs in Bayesian inverse problems. I was supported by the SIAM Student Travel Award. [slides]

• November 2022: I will transition to a postdoctoral position at The Oden Institute, under the supervision of J. Tinsley Oden and Omar Ghattas. I will be a part of the new M2dt (Multifaceted Mathematics of Digital Twins) center, a Multifaceted Mathematics Integrated Capability Center funded by the DOE's ASCR program. [news] [website]

• Oct. 2022: Our new paper (co-authored with Prashant, Tom, Dr. Oden & Omar) is now announced on arXiv! It aims to understand and improve the reliability of neural operators as surrogates of parametric PDEs in infinite-dimensional Bayesian inverse problems. [preprint]

• Jun. 2022: I gave a contributed talk at USNC/TAM 2022, in MS: Phase Field Study of Microstructures and Behaviors of Advanced Materials [website]. I presented results based on our SISC paper [journal] on a globally convergent modified Newton method for the direct minimization of the Ohata–Kawasaki energy, a phase field model of diblock copolymer self-assembly. [slides]

• Jun. 2022: Our new paper (co-authored by Ricardo, Josh, Fengyi, Youssef, Dr. Oden & Omar) is now announced on arXiv! In this paper, we present a Bayesian framework for calibrating the diblock copolymer self-assembly model under aleatoric uncertainties. The model calibration is solved using a likelihood-free inference method via measure transport in conjunction with summary statistics designs. We use expected information gain to measure the utility of summary statistics. [preprint]

• May 2022: I gave a contributed talk at EMI 2022, in MS: Physics-Based Data-Driven Modeling and Uncertainty Quantification in Computational Materials Science and Engineering [website]. I presented unpublished work on the Bayesian calibration of a diblock copolymer thin film self-assembly model using the power spectrum of microscopy image data. The paper containing the presented results is coming soon! [slides]