Welcome! I am a postdoctoral researcher at MIT, where I am affiliated with the Institute for Data, Systems, and Society (IDSS) and the Uncertainty Quantification Group. I obtained my PhD from Cambridge University, where I was advised by Richard Nickl.
In September 2023, I will join the mathematics department at Humboldt University Berlin as Juniorprofessor (assistant professor). Here's my CV. Email: svenwang [at] mit [dot] edu My research interests (i) Much of my research has been dedicated to mathematical statistics. I have worked on theoretical guarantees for many common inference procedures in complex and high-dimensional statistical models, related to:
|
Publications / preprints
[10] Infinite-dimensional diffusion models for function spaces (with J. Pidstrigach, Y. Marzouk and S. Reich). Preprint (2023).
[9] Accounting for stakes in democratic decisions (with B. Flanigan and A. Procaccia). Preprint (2023). Accepted at Foundations of Responsible Computing (FORC 2023, non-archival track).
[8] On free energy barriers in Gaussian priors and failure of MCMC for high-dimensional unimodal distributions (with A.S. Bandeira, A. Maillard and R. Nickl). Philosophical Transactions of the Royal Society A 381 (2023).
[7] Distortion under Public-Spirited Voting (with B. Flanigan and A. Procaccia). To appear in Economics and Computation (EC 2023).
[6] On minimax density estimation via measure transport (with Y. Marzouk). Preprint (2022).
[5] Wasserstein Distributionally Robust Gaussian Process Regression and Linear Inverse Problems (with X. Zhang, J. Blanchet, Y. Marzouk and V.A. Nguyen). Preprint (2022).
[4] Laplace priors and spatial inhomogeneity in Bayesian inverse problems (with S. Agapiou). To appear in Bernoulli.
[3] On polynomial-time computation of high-dimensional posterior measures by Langevin-type algorithms (with R. Nickl). Journal of the European Mathematical Society (2022). Link to EMS online publication.
[2] Convergence rates for Penalised Least Squares estimators in PDE-constrained regression problems (with R. Nickl and S. van de Geer). SIAM/ASA Journal of Uncertainty Quantification 8 (2020) 374-413.
[1] The nonparametric LAN expansion for discretely observed diffusions. Electronic Journal of Statistics 13 (2019) 1329-1358.
[A] Statistical inference and computation in PDE models. PhD thesis, University of Cambridge (2021).
[9] Accounting for stakes in democratic decisions (with B. Flanigan and A. Procaccia). Preprint (2023). Accepted at Foundations of Responsible Computing (FORC 2023, non-archival track).
[8] On free energy barriers in Gaussian priors and failure of MCMC for high-dimensional unimodal distributions (with A.S. Bandeira, A. Maillard and R. Nickl). Philosophical Transactions of the Royal Society A 381 (2023).
[7] Distortion under Public-Spirited Voting (with B. Flanigan and A. Procaccia). To appear in Economics and Computation (EC 2023).
[6] On minimax density estimation via measure transport (with Y. Marzouk). Preprint (2022).
[5] Wasserstein Distributionally Robust Gaussian Process Regression and Linear Inverse Problems (with X. Zhang, J. Blanchet, Y. Marzouk and V.A. Nguyen). Preprint (2022).
[4] Laplace priors and spatial inhomogeneity in Bayesian inverse problems (with S. Agapiou). To appear in Bernoulli.
[3] On polynomial-time computation of high-dimensional posterior measures by Langevin-type algorithms (with R. Nickl). Journal of the European Mathematical Society (2022). Link to EMS online publication.
[2] Convergence rates for Penalised Least Squares estimators in PDE-constrained regression problems (with R. Nickl and S. van de Geer). SIAM/ASA Journal of Uncertainty Quantification 8 (2020) 374-413.
[1] The nonparametric LAN expansion for discretely observed diffusions. Electronic Journal of Statistics 13 (2019) 1329-1358.
[A] Statistical inference and computation in PDE models. PhD thesis, University of Cambridge (2021).
Forthcoming articles
[11] Statistical consistency theory for sampling and density estimation via neural differential equations (with R. Ren, Y. Marzouk and J. Zech).
Invited conference talks (upcoming & past):
SIAM Conference on Computational Science and Engineering (2023/02)
Statistics for Stochastic Processes: SDEs, SPDEs and concentration of measure (2022/09)
ESI Vienna workshop "Statistical estimation and deep learning in UQ for PDEs" (2022/05, link to program)
AMS Sectional meeting (2022/05, link to program)
BNP workshop Cyprus (2022/04, link to program)
SIAM Conference on Uncertainty Quantification (2022/04)
Oberwolfach "Data Assimilation -- Mathematical Foundation and Applications" (2022/02)
BANFF "Statistical Aspects of Non-Linear Inverse Problems" (link to video)
Oberwolfach "Foundations of Bayesian Inference for Complex Statistical Models"
DPMMS Cambridge workshop "Mathematical and Statistical challenges in Uncertainty Quantification" (link to video)
Selected seminar talks: UMass Amherst (2022/03); WIAS, Berlin (2021/01, 2020/01); University of Chicago (2020/10); ETH Zurich YDSSZ seminar (2020/07); Caltech (2020/01); CREST Paris (2019/11); U of Sussex (2019/10)
SIAM Conference on Computational Science and Engineering (2023/02)
Statistics for Stochastic Processes: SDEs, SPDEs and concentration of measure (2022/09)
ESI Vienna workshop "Statistical estimation and deep learning in UQ for PDEs" (2022/05, link to program)
AMS Sectional meeting (2022/05, link to program)
BNP workshop Cyprus (2022/04, link to program)
SIAM Conference on Uncertainty Quantification (2022/04)
Oberwolfach "Data Assimilation -- Mathematical Foundation and Applications" (2022/02)
BANFF "Statistical Aspects of Non-Linear Inverse Problems" (link to video)
Oberwolfach "Foundations of Bayesian Inference for Complex Statistical Models"
DPMMS Cambridge workshop "Mathematical and Statistical challenges in Uncertainty Quantification" (link to video)
Selected seminar talks: UMass Amherst (2022/03); WIAS, Berlin (2021/01, 2020/01); University of Chicago (2020/10); ETH Zurich YDSSZ seminar (2020/07); Caltech (2020/01); CREST Paris (2019/11); U of Sussex (2019/10)
Music & Miscellanea
I love playing piano, singing and composing music. Some groups I've been lucky to be part of:
I also love being outdoors.
- Munich Bach choir
- Cambridge's Pembroke lieder scheme, where I studied with Joseph Middleton
- MIT Chamber Music Society
- Kammermusikkreis Unterwachingen
- Michael Schopper, my most wonderful mentor and singing teacher
- Here is me playing some late Schubert in 2019, recorded in Trinity College Chapel
I also love being outdoors.
- In 2015, I biked through 22 European countries for 6 months (with my very good friend Niklas from Berlin). We paired our trip with a fundraising project towards medical care for refugees from the war in Syria, which has completely disappeared from the news but is unfortunately still timely.
- In 2012, my friend Nick came up with the glorious idea of hiking across the alps with a donkey (named Thomas) from our German hometown Aschaffenburg to the Mediterranean Sea in Italy. It was truly life-changing for 16-year-old me. Thomas had a solar panel on his back, and we funded our trip by playing the accordion in the streets. We were (unintentionally) discovered by the Italian media -- they must have had a summer drought of stories to report about.