News & EventsDepartment Events
Events
-
Apr29
EVENT DETAILS
Title: Understanding the Effect of GCN Convolutions in Regression Tasks - NITMB Lecture
Speaker: Johannes Schmidt-Hieber, University of Twente
Special Note: This talk will be streamed from NITMB via zoom. Our usual colloquium room M416 will be used as a satellite room for viewing the stream. The talk can also be viewed individually on zoom using the zoom link below.
Abstract: Graph Convolutional Networks (GCNs) have become a pivotal method in machine learning for modeling functions over graphs. Despite their widespread success across various applications, their statistical properties (e.g. consistency, convergence rates) remain ill-characterized. To begin addressing this knowledge gap, in this paper, we provide a formal analysis of the impact of convolution operators on regression tasks over homophilic networks. Focusing on estimators based solely on neighborhood aggregation, we examine how two common convolutions - the original GCN and GraphSage convolutions - affect the learning error as a function of the neighborhood topology and the number of convolutional layers. We explicitly characterize the bias-variance trade-off incurred by GCNs as a function of the neighborhood size and identify specific graph topologies where convolution operators are less effective. Our theoretical findings are corroborated by synthetic experiments, and provide a start to a deeper quantitative understanding of convolutional effects in GCNs for offering rigorous guidelines for practitioners.
Joint work with Juntong Chen (Twente), Claire Donnat (U Chicago), and Olga Klopp (ESSEC Business School, Paris)Zoom: https://northwestern.zoom.us/s/93835110431
-----
To subscribe to the Applied Mathematics Colloquia List send a message to LISTSERV@LISTSERV.IT.NORTHWESTERN.EDU with the command:
SUBSCRIBE esam-seminar FirstName LastName
TIME Tuesday, April 29, 2025 at 11:15 AM - 12:15 PM
LOCATION M416, Technological Institute map it
CONTACT Ted Shaeffer ted.shaeffer@northwestern.edu EMAIL
CALENDAR McCormick-Engineering Sciences and Applied Mathematics (ESAM)
-
May6
EVENT DETAILS
Title: Improving Global Atmosphere Simulation in Earth System Models with Multiphysics Time Integration Methods
Speaker: Chris Vogl, Lawrence Livermore National Laboratory
Abstract: To reliably predict the frequency and severity of hurricanes, floods, droughts, and other weather-driven disasters, the efficient simulation of Earth system models is crucial. Such an endeavor poses a complex, multiphysics problem involving modeling, temporal and spatial discretization, and software implementation considerations. This work focuses on the time integration of the global atmosphere component, with an emphasis on the bulk atmosphere flow and cloud microphysics models. The nonhydrostatic bulk atmosphere flow models include acoustic waves that make the overall system numerically stiff. Our work has developed a model formulation that is amenable to an IMEX approach, where the acoustic waves are treated implicitly. The performance of both existing and customized additive Runge-Kutta methods is evaluated, with certain methods remaining stable at the hydrostatic timestep. Cloud microphysics models currently use first-order operator splitting to address the multiple timescales in the modeled physics, with at-best-first-order limiters required to keep quantities physical. Our work has shown that higher-order explicit, implicit, and IMEX Runge-Kutta methods with error-based adaptive timestep control are more efficient for the subset of cloud microphysics considered thus far.
Zoom: https://northwestern.zoom.us/j/99632064287
-----
To subscribe to the Applied Mathematics Colloquia List send a message to LISTSERV@LISTSERV.IT.NORTHWESTERN.EDU with the command:
SUBSCRIBE esam-seminar FirstName LastName
TIME Tuesday, May 6, 2025 at 11:15 AM - 12:15 PM
LOCATION M416, Technological Institute map it
CONTACT Ted Shaeffer ted.shaeffer@northwestern.edu EMAIL
CALENDAR McCormick-Engineering Sciences and Applied Mathematics (ESAM)
-
May12
EVENT DETAILS
Title: In Search of Internal Mental Models - Reiss Lecture
Speaker: Adrienne Fairhall, University of Washington
Abstract: How do we build the mental models that we use to perceive, navigate and reason about the world? How might these models be inferred from neural activity? I will describe experiments and analysis in collaboration with Beth Buffalo's lab to explore these questions in our closest relatives, nonhuman primates. In one example, we compare monkey and human behavior in a decision task, and analyze how subjects make use of visual information and feedback to infer a hidden rule, where the rule switches in an uncued fashion. We fit a suite of behavioral models and learn that while humans are close to optimal Bayesian agents, monkey behavior is better fit as reinforcement learning. This allows us to seek neural implementations of this internal belief update. Further, while rodent hippocampus famously encodes the animal's spatial location, we find evidence that hippocampus in the primate serves a more cognitive role.
Zoom: TBA
-----
To subscribe to the Applied Mathematics Colloquia List send a message to LISTSERV@LISTSERV.IT.NORTHWESTERN.EDU with the command:
SUBSCRIBE esam-seminar FirstName LastName
TIME Monday, May 12, 2025 at 4:00 PM - 5:00 PM
LOCATION Hive Room 2350, Ford Motor Company Engineering Design Center map it
CONTACT Ted Shaeffer ted.shaeffer@northwestern.edu EMAIL
CALENDAR McCormick-Engineering Sciences and Applied Mathematics (ESAM)
-
May13
EVENT DETAILS
Title: Decoding Neurons to Behavior in a Model Organism - Reiss Lecture
Speaker: Adrienne Fairhall, University of Washington
Abstract: The freshwater cnidarian Hydra is a fascinating model organism for neuroscience. It is transparent; new genetic lines allow one to image activity in both neurons and muscle cells; it exhibits a quite rich suite of behaviors; and it continually rebuilds itself. Hydra’s fairly simple physical structure as a two-layered fluid-filled hydrostat and the accessibility of information about neural and muscle activity open the possibility of a complete model of neural control of behavior. We have developed a biophysical and biomechanical model of Hydra's muscles and body that allows us to transform measured neural activity into behavior. We also propose a model for how the neural network rebuilds as the animal regenerates itself following bisection.
Zoom: https://northwestern.zoom.us/j/94136984398
-----
To subscribe to the Applied Mathematics Colloquia List send a message to LISTSERV@LISTSERV.IT.NORTHWESTERN.EDU with the command:
SUBSCRIBE esam-seminar FirstName LastName
TIME Tuesday, May 13, 2025 at 4:00 PM - 5:00 PM
LOCATION Hive Room 2350, Ford Motor Company Engineering Design Center map it
CONTACT Ted Shaeffer ted.shaeffer@northwestern.edu EMAIL
CALENDAR McCormick-Engineering Sciences and Applied Mathematics (ESAM)
-
May27
EVENT DETAILS
Title: Two Tales, One Resolution: Physics-Informed Test Time Scaling and Precondition
Speaker: Yiping Lu, Northwestern University
Abstract: In this talk, I will introduce a novel framework for physics-informed debiasing of machine learning estimators, which we call Simulation-Calibrated Scientific Machine Learning (SCaSML). This approach leverages the structure of physical models to achieve two key objectives:
Unbiased Predictions: It produces unbiased predictions even when the underlying machine learning predictor is biased.
Overcoming Dimensionality Challenges: It mitigates the curse of dimensionality that often affects high-dimensional estimators.The SCaSML paradigm integrates a (potentially) biased machine learning algorithm with a de-biasing procedure that is rigorously designed using numerical analysis and stochastic simulation. Our methodology aligns with recent advances in inference-time computation—similar to those seen in the large language model literature—demonstrating that additional computation can enhance ML estimates. Furthermore, we establish a surprising equivalence between our framework and another research direction that utilizes approximate (linearized) solvers to precondition iterative methods. This connection not only bridges two distinct areas of study but also offers new insights into improving estimation accuracy in complex, high-dimensional (PDE) settings.
Zoom: https://northwestern.zoom.us/j/94570889326
-----
To subscribe to the Applied Mathematics Colloquia List send a message to LISTSERV@LISTSERV.IT.NORTHWESTERN.EDU with the command:
SUBSCRIBE esam-seminar FirstName LastName
TIME Tuesday, May 27, 2025 at 11:15 AM - 12:15 PM
LOCATION M416, Technological Institute map it
CONTACT Ted Shaeffer ted.shaeffer@northwestern.edu EMAIL
CALENDAR McCormick-Engineering Sciences and Applied Mathematics (ESAM)
-
Jun15
EVENT DETAILSmore info
2024-2025 Commencement Ceremony
TIME Sunday, June 15, 2025
CONTACT Office of the Registrar nu-registrar@northwestern.edu EMAIL
CALENDAR University Academic Calendar
-
Jun16
EVENT DETAILS
McCormick School of Engineering PhD Hooding and Master's Degree Recognition Ceremony. The most up to date information can be found on our graduation webpage.
TIME Monday, June 16, 2025 at 9:00 AM - 11:00 AM
LOCATION 2705 Ashland Ave
CONTACT Northwestern Engineering Events northwestern-engineering-events@northwestern.edu EMAIL
CALENDAR McCormick School of Engineering and Applied Science
-
Jun16
EVENT DETAILSmore info
McCormick School of Engineering Undergraduate Convocation. The most up to date information can be found on our graduation webpage.
TIME Monday, June 16, 2025 at 2:00 PM - 4:00 PM
LOCATION 2705 Ashland Ave
CONTACT Northwestern Engineering Events northwestern-engineering-events@northwestern.edu EMAIL
CALENDAR McCormick School of Engineering and Applied Science