Benjamin J. Zhang

Benjamin J. Zhang

Auto-generated profile, based on publicly available information

Benjamin J. Zhang

Department of Aeronautics and Astronautics Center for Computational Science and Engineering Massachusetts Institute of Technology Cambridge MA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 January 2022 in Journal of Computational Physics
Verified
Journal Article
Closed access
BZBenjamin J. Zhang
Benjamin J. Zhang
TSTuhin Sahai
Tuhin Sahai
YMYoussef M. Marzouk
Youssef M. Marzouk

19 September 2022 in Statistics and Computing
Verified
Journal Article
Open access
PDF file
BZBenjamin J. Zhang
Benjamin J. Zhang
YMYoussef M. Marzouk
Youssef M. Marzouk
KSKonstantinos Spiliopoulos
Konstantinos Spiliopoulos
We introduce a novel geometry-informed irreversible perturbation that accelerates convergence of the Langevin algorithm for Bayesian computation. It is well documented that there exist perturbations to the Langevin dynamics that preserve its invariant measure while accelerating its convergence. Irreversible perturbations and reversible perturbations (such as Riemannian manifold Langevin dynamics (RMLD)) have separately been shown to improve the performance of Langevin samplers. We consider these two perturbations simultaneously by presenting a novel form of irreversible perturbation for RMLD that is informed by the underlying geometry. Through numerical examples, we show that this new irreversible perturbation can improve estimation performance over irreversible perturbations that do not take the geometry into account. Moreover we demonstrate that irreversible perturbations generally can be implemented in conjunction with the stochastic gradient version of the Langevin algorithm. Lastly, while continuous-time irreversible perturbations cannot impair the performance of a Langevin estimator, the situation can sometimes be more complicated when discretization is considered. To this end, we describe a discrete-time example in which irreversibility increases both the bias and variance of the resulting estimator.

18 July 2025 in Computer Methods in Applied Mechanics and Engineering
Verified
Journal Article
Open access
JBJeremiah Birrell
Jeremiah Birrell
MKMarkos A. Katsoulakis
Markos A. Katsoulakis
LRLuc Rey-Bellet
Luc Rey-Bellet
BZBenjamin J. Zhang
Benjamin J. Zhang
WZWei Zhu
Wei Zhu

28 September 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
KKKelvin Kan
Kelvin Kan
XLXingjian Li
Xingjian Li
BZBenjamin J. Zhang
Benjamin J. Zhang
TSTuhin Sahai
Tuhin Sahai
SOStanley Osher
Stanley Osher
MKMarkos A. Katsoulakis
Markos A. Katsoulakis
Despite their widespread use, training deep Transformers can be unstable. Layer normalization, a standard component, improves training stability, but its placement has often been ad hoc. In this paper, we conduct a principled study on the forward (hidden states) and backward (gradient) stability of Transformers under different layer normalization placements. By utilizing discrete-time optimal control theory, our results provide key insights into the training dynamics: whether training drives Transformers toward regular solutions or pathological behaviors. For forward stability, we derive explicit bounds on the growth of hidden states in trained Transformers. For backward stability, we analyze how layer normalization affects the backpropagation of gradients, thereby explaining the training dynamics of each layer normalization placement. Our analysis also guides the scaling of residual steps in Transformer blocks, where appropriate choices can further improve stability and performance. Our numerical results corroborate our theoretical findings. Beyond these results, our framework provides a principled way to sanity-check the stability of Transformers under new architectural modifications, offering guidance for future designs.

7 February 2022 in arXiv
Verified
Preprint
Open access
PDF file
BZBenjamin Zhang
Benjamin Zhang
TSTuhin Sahai
Tuhin Sahai
YMYoussef Marzouk
Youssef Marzouk
We exploit the relationship between the stochastic Koopman operator and the Kolmogorov backward equation to construct importance sampling schemes for stochastic differential equations. Specifically, we propose using eigenfunctions of the stochastic Koopman operator to approximate the Doob transform for an observable of interest (e.g., associated with a rare event) which in turn yields an approximation of the corresponding zero-variance importance sampling estimator. Our approach is broadly applicable and systematic, treating non-normal systems, non-gradient systems, and systems with oscillatory dynamics or rank-deficient noise in a common framework. In nonlinear settings where the stochastic Koopman eigenfunctions cannot be derived analytically, we use dynamic mode decomposition (DMD) methods to compute them numerically, but the framework is agnostic to the particular numerical method employed. Numerical experiments demonstrate that even coarse approximations of a few eigenfunctions, where the latter are built from non-rare trajectories, can produce effective importance sampling schemes for rare events.

1 September 2022 in arXiv
Verified
Preprint
Open access
PDF file
BZBenjamin J. Zhang
Benjamin J. Zhang
YMYoussef M. Marzouk
Youssef M. Marzouk
KSKonstantinos Spiliopoulos
Konstantinos Spiliopoulos
We introduce a novel geometry-informed irreversible perturbation that accelerates convergence of the Langevin algorithm for Bayesian computation. It is well documented that there exist perturbations to the Langevin dynamics that preserve its invariant measure while accelerating its convergence. Irreversible perturbations and reversible perturbations (such as Riemannian manifold Langevin dynamics (RMLD)) have separately been shown to improve the performance of Langevin samplers. We consider these two perturbations simultaneously by presenting a novel form of irreversible perturbation for RMLD that is informed by the underlying geometry. Through numerical examples, we show that this new irreversible perturbation can improve estimation performance over irreversible perturbations that do not take the geometry into account. Moreover we demonstrate that irreversible perturbations generally can be implemented in conjunction with the stochastic gradient version of the Langevin algorithm. Lastly, while continuous-time irreversible perturbations cannot impair the performance of a Langevin estimator, the situation can sometimes be more complicated when discretization is considered. To this end, we describe a discrete-time example in which irreversibility increases both the bias and variance of the resulting estimator.