Loading...
Organizing Committee
George Mason University
Rice University
Brown University
Tufts University
University of California, Merced
Abstract
Optimization is ubiquitous in science and engineering. In engineering, one may have to tackle optimization problems constrained by large systems, and in machine learning, one tackles optimization problems with large-scale datasets. These problems can have significant stochastic components, this could be algorithmic (stochastic gradient) or unknown parameters in underlying systems which could be modeled as random variables. The goal of the workshop is to bring together researchers working in optimization, scientific machine learning, and computational science to address pressing challenges at the intersection of engineering, data science, and scientific computing. Topics of interest may include new methodologies for simulation-constrained optimization, optimization under uncertainty, optimization in distributed computing environments, reduced order modeling, efficient training of networks, and uncertainty quantification for machine learning. A wide range of scientific applications will be considered.
Loading...
Loading participant list in background...
Workshop Schedule
Monday, April 13, 2026
8:50 - 9:00 AM EDT
Welcome
11th Floor Lecture Hall
Brendan Hassett, ICERM/Brown University
9:00 - 9:45 AM EDT
Adaptive methods and neural ODEs in machine learning
11th Floor Lecture Hall
Speaker
Michael Hintermüller, Weierstrass Institute
Session Chair
Harbir Antil, George Mason University
Abstract
In this talk, we present a novel layerwise adaptive construction method for neural network architectures, which is based on a goal-oriented dual-weighted residual technique for the optimal control of neural differential equations. This leads to an ordinary differential equation constrained optimization problem with controls acting as coefficients and a specific loss function. We discuss our approach on the basis of a DG(0) Galerkin discretization of the neural ODE, leading to an explicit Euler time marching scheme. For the optimization we use steepest descent. Finally, we apply our method to the construction of neural networks for the classification of data sets, where we present results for a selection of well known examples from the literature.
10:00 - 10:30 AM EDT
Coffee Break
11th Floor Collaborative Space
10:30 - 11:15 AM EDT
The latent variable proximal point algorithm for variational problems with inequality constraints
11th Floor Lecture Hall
Speaker
Patrick Farrell, University of Oxford
Session Chair
Harbir Antil, George Mason University
Abstract
The latent variable proximal point (LVPP) algorithm is a new framework for solving infinite-dimensional variational problems with inequality constraints. The algorithm is a saddle point reformulation of the Bregman proximal point algorithm. At the continuous level, the two formulations are equivalent, but the saddle point formulation is more amenable to discretisation.
LVPP yields numerical methods with observed mesh-independence for obstacle problems, contact, fracture, plasticity, and others besides. In many cases this mesh independence is achieved for the first time. The framework also extends to more complex constraints, providing means to enforce convexity in the Monge-Ampère equation and gracefully handling quasi-variational inequalities, where the underlying constraint depends implicitly on the unknown solution.
In this talk we describe the LVPP algorithm in a general form and apply it to a number of problems from across mathematics.
11:30 AM - 12:15 PM EDT
Optimization-Ready Surrogates with Derivative-Informed Learning
11th Floor Lecture Hall
Speaker
Thomas O'Leary-Roseberry, Ohio State University
Session Chair
Harbir Antil, George Mason University
Abstract
Modern decision-making in complex physical and engineered systems increasingly requires quantifying high-dimensional uncertainty and solving risk-averse optimization problems under uncertainty, often in near real time. Operator learning has emerged as a promising framework for constructing scalable surrogate models in this setting. Yet approximation errors in learned operators can significantly degrade the quality of downstream decisions arising from inverse problems and optimization tasks.
In this talk, we present a framework that formulates operator learning through the lens of downstream optimization objectives. By deriving a priori error bounds for inverse problems and optimization under uncertainty, we motivate learning formulations that explicitly penalize errors in the derivatives of input–output maps. We illustrate the effectiveness of these ideas on applications including structural health monitoring, shape optimization, and fluid flow control. The resulting methods yield improved accuracy in surrogate-based optimization and consistently stronger empirical performance in related statistical learning tasks.
12:30 - 2:30 PM EDT
Lunch/Free Time
2:30 - 3:15 PM EDT
Programmable multifunctional materials and structures: Design, realization, and validation
11th Floor Lecture Hall
Speaker
Xiaojia Shelly Zhang, University of Illinois Urbana-Champaign
Session Chair
Elizabeth Newman, Tufts University
Abstract
Programmable materials capable of adaptive, controllable, and stimuli-responsive behaviors are enabling a new generation of functional devices in soft robotics, bio-integrated systems, and smart structures. Realizing their full potential requires synergistic integration of computational design and advanced fabrication strategies. In this talk, we present strategies that bridge the designs and additive manufacturing to enable the physical realization of programmable materials with multiphysics interactions. Computational designs with various programmable properties are generated by multi-material, multi-physics topology optimization approaches that optimize geometry, material distribution, and internal architectures to achieve targeted mechanical and actuation responses under large deformations. To fabricate the obtained designs with complex geometries and heterogeneous material/phase distributions, we introduce various techniques, including a hybrid fabrication method, tailored path generation methods, and 4D direct ink writing (DIW) techniques. The latter methods align local material orientations and printing paths with optimized fields, allowing the physical realization of designs with continuous distributions. We demonstrate this integration through experimental validation of diverse functional prototypes, including soft multimaterial metastructures, magnetically responsive materials with spatially varying magnetization patterns, and liquid crystal elastomer (LCE) systems capable of complex temperature-driven morphing behaviors. These developments represent a step toward programmable matter whose properties and responses can be tailored and reconfigured by design and manufacturing for applications in responsive actuation, reconfigurable devices, and intelligent system.
3:30 - 4:00 PM EDT
Coffee Break
11th Floor Collaborative Space
4:00 - 4:45 PM EDT
Optimality conditions for a maximum probability problem in optimal control
11th Floor Lecture Hall
Speaker
Caroline Geiersbach, University of Klagenfurt
Session Chair
Elizabeth Newman, Tufts University
Abstract
In this talk, we consider a linear parabolic partial differential equation in which the control appears as a distributed source term and uncertainty acts in a structured way on the right-hand side. The objective is find a control such that some outcome of the random state occurs with maximal probability. The outcome can be enforced pointwise, for instance when requiring the state to remain bounded at every point in time and space. This type of model offers a compromise between the robust "lookback" problem and the problem of maximizing on average. Probability functions are attractive because of their interpretability, but are generally nonsmooth and nonconvex. In the case of Gaussian-like noise, the properties of the underlying control-to-state operator allow us to obtain a fully explicit formula for the Clarke subdifferential of the probability function. The formula depends on certain solutions to an adjoint equation with measures appearing in source terms. We use this formula to compute solutions in a numerical example. (Joint work with Hasnaa Zidani and Wim van Ackooij.)
5:00 - 6:30 PM EDT
Reception
11th Floor Collaborative Space
Tuesday, April 14, 2026
9:00 - 9:45 AM EDT
Space-time convergence of a physics-compatible discrete scheme for optimal control of Maxwell's equations
11th Floor Lecture Hall
Speaker
Denis Ridzal, Sandia National Laboratorie
Session Chair
Matthias Heinkenschloss, Rice University
Abstract
We formulate electromagnetic cloaking as an optimal control problem constrained by the full time-dependent Maxwell’s equations. We seek a spatially localized current density that destructively interferes with a signal generated by a localized source, to form a region of arbitrary shape with near-zero electromagnetic energy. Previous work addresses a related problem by assuming a factorization of the control current into a temporally constant function and a spatially constant function. We show that this factorization leads to suboptimal results. We discretize our general control problem using the Crank-Nicolson method with Nedelec and Raviart-Thomas finite elements. The discretization is structure-preserving as well as energy- and charge-conserving. We prove that the discretization of the control problem is well-posed, and we derive its optimality conditions. We demonstrate active cloaking through numerical simulations.
10:00 - 10:30 AM EDT
Coffee Break
11th Floor Collaborative Space
10:30 - 11:15 AM EDT
An Inexact Trust-Region Algorithm for Nonsmooth Risk-Averse PDE-Constrained Optimization
11th Floor Lecture Hall
Speaker
Drew Kouri, Sandia National Laboratories
Session Chair
Matthias Heinkenschloss, Rice University
Abstract
Many practical problems require the optimization of systems of PDEs with uncertain inputs such as noisy problem data, unknown operating conditions, and unverifiable modeling assumptions. In this talk, we formulate these problems as infinite-dimensional, risk-averse stochastic programs for which we minimize a quantification of risk associated with the system performance. For many popular risk measures, the resulting risk-averse objective function is not differentiable, significantly complicating the numerical solution of the optimization problem. Unfortunately, traditional methods for nonsmooth optimization converge slowly (e.g., sublinearly) and consequently are often intractable for problems in which the objective function and any derivative information is expensive to evaluate. To address this challenge, we introduce a novel trust-region algorithm for solving large-scale nonsmooth risk-averse optimization problems. This algorithm is motivated by the primal-dual risk minimization algorithm and employs smooth approximate risk measures at each iteration. In addition, this algorithm permits and rigorously controls inexact objective function value and derivative (when available) computations, enabling the use of inexpensive approximations such as adaptive discretizations. We discuss convergence of the algorithm under mild assumptions and demonstrate its efficiency on various examples from PDE-constrained optimization.
11:30 AM - 12:15 PM EDT
Statistical Limit Theorems for PDE-Constrained Optimization under Uncertainty
11th Floor Lecture Hall
Speaker
Johannes Milz, Georgia Institute of Technology
Session Chair
Matthias Heinkenschloss, Rice University
Abstract
Many solution approaches for PDE-constrained optimization under uncertainty rely on Monte Carlo sampling, replacing expectations by finite-sample averages. The resulting optimal values and controls are random and subject to sampling variability. We establish statistical limit theorems that characterize the large-sample fluctuations of these sample-based estimators. These results facilitate practical uncertainty quantification: they give confidence regions for optimal values and controls. They also help guide decisions about sample sizes and compute resource needs. While sample-based optimal values often have Gaussian limits, solution fluctuations are typically non-Gaussian in the presence of control constraints or other nonsmooth regularizers. We illustrate the theory in applications to risk-neutral PDE-constrained optimization, stochastic optimal control in continuous time, and learning problems posed in Hilbert spaces.
This is joint work with Olena Melnikov.
12:30 - 2:30 PM EDT
Lunch/Free Time
2:30 - 3:00 PM EDT
Poster Blitz
Lightning Talks - 11th Floor Lecture Hall
Session Chair
Denis Ridzal, Sandia National Laboratorie
3:00 - 4:00 PM EDT
4:00 - 4:45 PM EDT
Optimization-Based Simulation of High-Speed Flows
11th Floor Lecture Hall
Speaker
Matthew Zahr, University of Notre Dame
Session Chair
Denis Ridzal, Sandia National Laboratorie
Abstract
In this talk, we present a class of optimization-based numerical methods for the simulation of high-speed compressible flows. The proposed framework leverages nonlinear finite element manifolds that adapt the approximation space to the most salient flow features, including boundary layers, shock waves, and rarefaction regions. By embedding problem-specific structure directly into the trial space, these methods enhance accuracy and efficiency, particularly in regimes characterized by strong gradients and discontinuities. Both the linear coefficients and the nonlinear parameters defining the approximation manifold are determined through the solution of an optimization problem. Numerical results demonstrate the potential of this approach to deliver highly accurate solutions on extremely coarse grids.
Wednesday, April 15, 2026
9:00 - 9:45 AM EDT
RALES: A Risk-Averse Augmented Lagrangian Equality-constrained SQP Method for Optimization with General Constraints
11th Floor Lecture Hall
Speaker
Thomas Surowiec, Simula Research Laboratory
Session Chair
Brendan Keith, Brown University
Abstract
Recent advances in augmented Lagrangian methods for infinite-dimensional constrained optimization have significantly broadened this classical algorithmic framework. However, cutting-edge approaches like ALESQP were not designed to handle the nonsmooth composite objective functions typically found in risk-averse stochastic optimization models. Conversely, methods such as the primal-dual risk minimization (PD-Risk) algorithm, developed for minimizing structured nonsmooth risk measures, do not account for simulation-based equality constraints or complex geometric constraints. We remedy this gap by fusing PD-Risk with ALESQP to introduce a new method: RALES. RALES is a provably convergent algorithm designed to solve risk-averse optimization problems that incorporate both simulation-based equality constraints and general convex set constraints. The method specifically addresses the computational challenges posed by nonsmooth coherent risk measures, such as Average Value-at-Risk (AVaR). It achieves this by utilizing a risk-averse augmented Lagrangian functional that maintains Fréchet differentiability via epiregularization of the nonsmooth functionals using the previous dual variables in the regularization terms. By integrating this functional with a composite-step trust-region SQP solver, RALES effectively decomposes the problem into manageable components. Simulation-based constraints are handled via inexact, matrix-free linear system solves, while general convex constraints are managed through the augmented Lagrangian. A distinguishing feature of the algorithm is its use of a decoupled penalty parameter update scheme (adapted from the ALESQP framework) during an initial calibration phase of K iterations. This allows the framework to adaptively learn appropriate scalings for heterogeneous sets of constraints. We provide a rigorous convergence analysis for minimizers and dual variables in an ideal setting, and stationary points in a more general, computationally relevant setting. Finally, we illustrate the method's performance and viability through numerical experiments on high-dimensional problems.
10:00 - 10:30 AM EDT
Coffee Break
11th Floor Collaborative Space
10:30 - 11:15 AM EDT
Optimization with Superquantile Constraints: A Fast Computational Approach
11th Floor Lecture Hall
Speaker
Ying Cui, University of California, Berkeley
Session Chair
Brendan Keith, Brown University
Abstract
We present an efficient and scalable second-order computational framework for solving large-scale optimization problems with superquantile constraints. Unlike empirical risk models, superquantile models have non-separable constraints that make typical first-order algorithms difficult to scale. We address the challenge by adopting a hybrid of the second-order semismooth Newton method and the augmented Lagrangian method, which takes advantage of the structured sparsity brought by the risk sensitive measures. The key to make the proposed computational framework scalable in terms of the number of training data is that the matrix-vector multiplication in solving the resulting Newton system can be computed in a reduced space due to the aforementioned sparsity. The computational cost per iteration for the Newton method is similar or even smaller than that of the first-order method. Our developed solver is expected to help improve the reliability and accuracy of statistical estimation and prediction, as well as control the risk of adverse events for safety-critical applications.
11:30 AM - 12:15 PM EDT
Tensor train solution to optimization problems constrained by differential equations under uncertainty
11th Floor Lecture Hall
Speaker
Akwum Onwunta, Lehigh University
Session Chair
Brendan Keith, Brown University
Abstract
We present an algorithm to solve high-dimensional risk-averse optimization problems governed by differential equations (ODEs and/or partial differential equations [PDEs]) under uncertainty. As an example, we focus on the so-called Conditional Value -at-Risk (CVaR), but the approach is equally applicable to other coherent risk measures. Both the full and reduced space formulations are considered. The algorithm is based on low rank tensor approximations of random fields discretized using stochastic collocation. To avoid nonsmoothness of the objective function underpinning the CVaR, we propose an adaptive strategy to select the width parameter of the smoothed CVaR to balance the smoothing and tensor approximation errors. Moreover, unbiased Monte Carlo CVaR estimate can be computed by using the smoothed CVaR as a control variate. To accelerate the computations, we introduce an efficient preconditioner for the Karush–Kuhn–Tucker (KKT) system in the full space formulation. We present numerical experiments to demonstrate that the proposed method enables accurate CVaR optimization constrained by large-scale discretized systems.
12:25 - 12:30 PM EDT
Group Photo (Immediately After Talk)
11th Floor Lecture Hall
12:30 - 2:30 PM EDT
Lunch/Free Time
2:30 - 3:00 PM EDT
Poster Blitz
Lightning Talks - 11th Floor Lecture Hall
Session Chair
Thomas Surowiec, Simula Research Laboratory
3:00 - 4:00 PM EDT
4:00 - 4:45 PM EDT
Uncertainty in Uncertainty and Rockafellian Relaxation
11th Floor Lecture Hall
Speaker
Sean Carney, Union College
Session Chair
Thomas Surowiec, Simula Research Laboratory
Abstract
Uncertainty in physical systems is modeled with random variables, however, in practice there may be some nontrivial ambiguity in the underlying probability distribution from which they are sampled. As stochastic optimal control problems are known to be ill-conditioned to perturbations in the sampling distribution, we describe an analytic framework that is better conditioned to such “meta-uncertainties” and conclude with numerical examples.
Thursday, April 16, 2026
9:00 - 9:45 AM EDT
Towards Digital Twin for Design of Experiments with Moving Sensor
11th Floor Lecture Hall
Speaker
Sven Leyffer, Argonne National Laboratory
Session Chair
Michael Hintermüller, Weierstrass Institute
Abstract
Digital twins are information constructs that mimic physical systems, are dynamically updated from data of their physical twin, have predictive capabilities, and inform decisions. We optimize the path of a mobile sensor to minimize the posterior uncertainty of a Bayesian inverse problem. Along its path, the sensor continuously takes measurements of the state, which is a physical quantity modeled as the solution of a partial differential equation (PDE) with uncertain parameters. We derive the closed-form expression of the posterior covariance matrix of the model parameters as a function of the path, and formulate the optimal experimental design problem for minimizing the posterior's uncertainty. We discretize the problem such that the cost function remains consistent under temporal refinement. Our formulation allows the addition of constraints that model the dynamics of the sensor platform and ensure that the path avoids obstacles and remains physically interpretable through a control parameterization. We present computational results for a convection-diffusion equation with unknown initial condition.
10:00 - 10:30 AM EDT
Coffee Break
11th Floor Collaborative Space
10:30 - 11:15 AM EDT
Path-OED: optimal path design for mobile sensors
11th Floor Lecture Hall
Speaker
Alen Alexanderian, North Caroline State University
Session Chair
Michael Hintermüller, Weierstrass Institute
Abstract
We consider infinite-dimensional Bayesian linear inverse problems constrained by PDEs where data are collected via moving sensors. In this context, we develop a framework for finding sensor paths that optimize the posterior uncertainty in a prediction functional of the inversion parameters. The mathematical formulations are first established in an infinite-dimensional setting. Subsequently, we consider discretization of the problem and present a scalable computational framework for finding optimal sensor paths. We also present illustrative numerical results, within the context of a model inverse problems governed by an advection diffusion equation.
11:30 AM - 12:15 PM EDT
End-to-End Training of High-Dimensional Optimal Control with Implicit Hamiltonians via Jacobian-Free Backpropagation
11th Floor Lecture Hall
Speaker
Deepanshu Verma, Clemson University
Session Chair
Michael Hintermüller, Weierstrass Institute
Abstract
Value function methods have proven effective for learning high-dimensional optimal feedback controllers, but they fundamentally rely on solving the Hamiltonian maximization problem in closed form. Many practical systems, from space shuttle reentry to multi-agent swarm dynamics, involve Hamiltonians that admit no such explicit solution, limiting the applicability of existing methods.
This talk presents a framework that overcomes this bottleneck by parameterizing the value function in an implicit neural network and training via Jacobian-Free Backpropagation (JFB). The approach preserves the structural advantages of value function parameterization: optimal controls are recovered using feedback form through the connection between Pontryagin's Maximum Principle and dynamic programming, enforcing physical consistency by construction. On the theoretical side, JFB is shown to converge to the stationary points of the expected optimal control objective. Empirically, we demonstrate that the framework is effective for high-dimensional problems, including multi-agent optimal consumption, swarm-based quadrotor and bicycle control.
12:30 - 2:30 PM EDT
Lunch/Free Time
2:30 - 3:15 PM EDT
TBA
11th Floor Lecture Hall
Speaker
Eric Cyr, Sandia National Laboratories
Session Chair
Caroline Geiersbach, University of Klagenfurt
3:30 - 4:00 PM EDT
Coffee Break
11th Floor Collaborative Space
4:00 - 4:45 PM EDT
Risk-averse density-based topology optimization
11th Floor Lecture Hall
Speaker
Boyan Lazarov, Lawrence Livermore National Laboratory
Session Chair
Caroline Geiersbach, University of Klagenfurt
Abstract
This talk presents a risk-averse topology optimization of linear elastic structures in the presence of uncertainty. We consider density-based formulations, in which the design is described by a continuous density field that takes values near one in solid regions and near zero in voids. The objective is to distribute a prescribed amount of material within a fixed design domain to minimize a performance metric of the design (e.g., compliance). The existence of solutions is ensured by regularization, which smooths the density field but introduces intermediate (“gray”) regions that require further processing. To obtain crisp, manufacturable designs, we often apply a Heaviside projection and incorporate erosion and dilation, yielding the classical robust formulation that optimizes against the worst-case realization of geometric perturbations. Even though effective for simple uncertainties, worst-case formulation becomes computationally impractical and often overly conservative when uncertainty stems from loading variability or rare failure scenarios such as support loss or localized damage. To address this, we replace the worst-case design with tail-sensitive risk measures, such as Conditional Value-at-Risk (CVaR) or Entropic Value-at-Risk (EVaR), which control the contribution of adverse events while preserving design freedom. For computationally efficient evaluation of CVaR and its sensitivities, we leverage its dual risk-envelope representation and couple it with importance sampling to concentrate simulations on the most influential tail realizations. The approach is demonstrated on 2D benchmarks and large-scale 3D problems.
Friday, April 17, 2026
9:00 - 9:45 AM EDT
Inference of constitutive relations from data through integration of PDE with neural networks
11th Floor Lecture Hall
Speaker
Georg Stadler, Courant Institute, New York University
Session Chair
Harbir Antil, George Mason University
Abstract
I will present a general framework for inferring constitutive relations from data, with an emphasis on compressible and incompressible fluid mechanics. The method relies on expressing isotropic constitutive laws as scalar functions of the principal invariants of the strain-rate tensor. To ensure that the learned models are well-posed, we infer a convex potential of dissipation rather than directly learning bulk and shear viscosity functions. Moreover, to avoid imposing a specific functional form for the constitutive law, we represent the dissipation potential using neural networks trained on data. In particular, we employ input-convex neural networks to enforce convexity. By combining PyTorch with the finite element library Firedrake, we embed the governing PDE into the training procedure, which leads to a PDE-constrained optimization problem for the network parameters. This strategy enables us to infer the constitutive relation from velocity data, as opposed to stress measurements. We apply this approach to learn concentration-dependent compressible constitutive laws in Lagrangian sea ice flow models that involve many interacting Lagrangian ice floes. This is a joint work with Gonzalo G. De Diego (NYU).
10:00 - 10:45 AM EDT
Topology Optimization with Embedded Interfaces: Methods, Multiphysics, and Emerging Applications
11th Floor Lecture Hall
Speaker
Niels Aage, Technical University of Denmark, Department of Civil and Mechanical Engingeering
Session Chair
Harbir Antil, George Mason University
Abstract
Reliable treatment of internal boundary conditions remains a central challenge in structural optimization, particularly for strongly coupled multiphysics systems. This talk presents recent methodological developments in density‑based topology optimization and cut‑element formulations, with emphasis on applications involving vibro‑acoustics with viscothermal effects, thermofluidics, fluid–structure interaction and surgical procedures. Special attention is given to computational techniques that ensure robust and consistent representations of complex, embedded interfaces within the design domain. These capabilities are shown to be crucial not only in classical multiphysics settings but also in emerging applications such as optimization of surgical procedures, where large interface deformations must be handled accurately. Through selected examples, it is demonstrated that such interface‑aware strategies are essential for solving demanding multiphysics design problems and that they open new conceptual and computational perspectives for advanced engineering and bio‑inspired applications.
11:00 - 11:30 AM EDT
Coffee Break
11th Floor Collaborative Space
11:30 AM - 12:15 PM EDT
An online, adaptive finite element method for problems in nonsmooth PDE-constrained optimization
11th Floor Lecture Hall
Speaker
Robert Baraldi, Sandia National Laboratories
Session Chair
Harbir Antil, George Mason University
Abstract
We present a trust-region-based adaptive finite-element framework to numerically
solve a class of nonsmooth PDE-constrained optimization problems. In particular, we consider the class of problems whose objective function is the sum of a smooth, possibly nonconvex, function and a nonsmooth convex function. Our method combines the robustness of inexact trust-region algorithms for nonsmooth problems with the efficiency of adaptive finite-element discretizations. Starting from a coarse mesh, the algorithm automatically refines the discretization based on reliable a posteriori error estimators for both the state and adjoint equations, systematically controlling the accuracy of the computed objective function value and gradient. This adaptive mechanism enables efficient resolution of localized phenomena and sparsity structures in the state and control variables. We demonstrate the effectiveness of our algorithm through numerical experiments on control and topology optimization examples.
Application Information
The deadline to submit an application to this program has passed.
Your Visit to ICERM
- ICERM Facilities
- ICERM is located on the 10th & 11th floors of 121 South Main Street in Providence, Rhode Island. ICERM's business hours are 8:30am - 5:00pm during this event. See our facilities page for more info.
- Traveling to ICERM
- ICERM is located at Brown University in Providence, Rhode Island. Providence's T.F. Green Airport (15 minutes south) and Boston's Logan Airport (1 hour north) are the closest airports. Providence is also on Amtrak's Northeast Corridor. See our travel page for full directions.
- Lodging
- Confirmed participants will receive an email to book a room at our preferred hotel, the Hampton Inn & Suites Providence Downtown. Contact programstaff@icerm.brown.edu with questions.Only use hotel links from ICERM’s official emails or this page. ICERM never uses third-party booking vendors.
- Childcare/Schools
- Those traveling with family can contact housing@icerm.brown.edu for childcare and school information.
- Technology Resources
- Wireless internet access and wireless printing are available for all ICERM visitors. Eduroam is supported. Thin clients provide access to web browsers, SSH terminals, and printing. See our Technology Resources page for setup info.
- Accessibility
- For special services, accommodations, or assistance, please contact accessibility@icerm.brown.edu in advance. Thank you.
- Discrimination and Harassment Policy
- ICERM is committed to a safe, professional, and welcoming environment. Brown University's Code of Conduct and related policies apply to all ICERM participants. Contact the ICERM Director or Associate Director Jenna Sousa for concerns.
- Fundamental Research
- ICERM programs promote Fundamental Research and mathematical education. Participants involved in sensitive or proprietary work should be aware of U.S. export control restrictions.
- Exploring Providence
- Providence offers world-class dining and attractions. Visit our Explore Providence page to see what’s near ICERM.
Visa Information
Contact visa@icerm.brown.edu for assistance.
- Eligible to be reimbursed
- B-1 or Visa Waiver Business (WB)
- Ineligible to be reimbursed
- B-2 or Visa Waiver Tourist (WT)
- Already in the US?
F-1 and J-1 not sponsored by ICERM: need to obtain a letter approving reimbursement from the International Office of your home institution prior to travel.
H-1B holders do not need a letter of approval.
All other visas: alert ICERM staff immediately about your situation.
ICERM does not reimburse visa fees. This chart is to inform visitors whether the visa they enter the US on allows them to receive reimbursement for the items outlined in their invitation letter.
Financial Support
This section is for general purposes only and does not indicate that all attendees receive funding. Please refer to your personalized invitation to review your offer.
- ORCID iD
- As this program is funded by the National Science Foundation (NSF), ICERM is required to collect your ORCID iD if you are receiving funding to attend this program. Be sure to add your ORCID iD to your Cube profile as soon as possible to avoid delaying your reimbursement.
- Acceptable Costs
- 1 roundtrip between your home institute and ICERM
- Flights on U.S. or E.U. airlines – economy class to either Providence airport (PVD) or Boston airport (BOS)
- Ground Transportation to and from airports and ICERM.
- Unacceptable Costs
- Flights on non-U.S. or non-E.U. airlines
- Flights on U.K. airlines
- Seats in economy plus, business class, or first class
- Change ticket fees of any kind
- Multi-use bus passes
- Meals or incidentals
- Advance Approval Required
- Personal car travel to ICERM from outside New England
- Multiple-destination plane ticket; does not include layovers to reach ICERM
- Arriving or departing from ICERM more than a day before or day after the program
- Multiple trips to ICERM
- Rental car to/from ICERM
- Flights on a Swiss, Japanese, or Australian airline
- Arriving or departing from airport other than PVD/BOS or home institution’s local airport
- 2 one-way plane tickets to create a roundtrip (often purchased from Expedia, Orbitz, etc.)
- Travel Maximum Contributions
- New England: $350
- Other contiguous US: $850
- Asia & Oceania: $2,000
- All other locations: $1,500
- Note these rates were updated in Spring 2023 and supersede any prior invitation rates.
- Reimbursement Requests
Request Reimbursement with Cube
Refer to the back of your ID badge for more information. Checklists are available at the front desk and in the Reimbursement section of Cube.
- Reimbursement Tips
- Scanned original receipts are required for all expenses
- Airfare receipt must show full itinerary and payment
- ICERM does not offer per diem or meal reimbursement
- Allowable mileage is reimbursed at prevailing IRS Business Rate and trip documented via Google Maps PDF
- Keep all documentation until you receive your reimbursement!
- Reimbursement Timing
6–8 weeks after all documentation is sent to ICERM. All reimbursements are reviewed by multiple Brown University offices.
- Reimbursement Deadline
Submissions must be received within 30 days of ICERM departure to avoid applicable taxes. No submissions are accepted after six months.
