mds22

Optimization in Machine Learning and Data Science

Optimization plays a central role in machine learning by providing tools that formulate and solve computational problems.

MDS22 Will Showcase the Latest Advances in the Mathematics of Data Science in San Diego (and Online)

This September, MDS22 will provide a unique opportunity to aggregate data science advances and exchange ideas.

Panel Discussion at MDS22 Prepares Students and Early-Career Researchers for the Workforce

At MDS22, a panel of researchers from academia and industry shared their career experiences and perspectives.

MDS22 Panel Addresses the Complex Interplay Between Academia and Industry

Industrial researchers discussed the beneficial relationship between academia and industry during a MDS22 panel.

Photos from the 2022 SIAM Conference on Mathematics of Data Science

View a selection of photos from the 2022 SIAM Conference on Mathematics of Data Science, including shots of...

Broader Engagement Program Expands to the 2022 SIAM Conference on Mathematics of Data Science

The Broader Engagement program promotes mentorship, networking, and other activities to foster a sense of community.

Learning to Optimize: Where Deep Learning Meets Optimization and Inverse Problems

The "learning to optimize” paradigm leverages machine learning to accelerate the discovery of optimization methods.

Identifiability and Sensitivity Analysis for Bayesian Parameter Estimation in Systems Biology

An ongoing challenge in systems biology is the estimation of model parameters to constrain a model against...

Preparing for the MDS22 Career Fair

A preparation guide for the MDS22 Career Fair on September 29 is available here.

MDS22 Prize Spotlight: Weijie Su

Congratulations to Weijie Su, the recipient of the 2022 SIAM Activity Group on Data Science Early Career Prize!

Fokker-Planck Dynamics Improve Multiscale Imaging of Head and Neck Tumors

At MDS22, Kyle Lafata introduced novel data-driven strategies that model the appearance and behavior of head and...

Data Assimilation and Uncertainty Quantification Techniques Improve Hemodynamic Models

Sahani Pathiraja investigated the utility of sequential data assimilation methods for hemodynamic modeling at...

Functional ANOVA Method Models the Recovery of Photosynthetic Activity After Mendocino Wildfire

At MDS22, Manju Johny used data to better understand post-wildfire vegetation levels in the Mendocino National Forest.

Mathematical Model Optimizes Shoe Midsoles for Peak Running Performance

At MDS22, Sarah Fay discussed a recent partnership with Adidas to optimize 3D-printed shoe midsoles for peak running...

Data Assimilation with Ensemble Kalman Filtering Advances Earthquake Forecasting Methods

At MDS22, Femke Vossepoel utilized data assimilation techniques with an ensemble Kalman filter to improve...

Data Assimilation Framework Sheds Light on Ventilator-induced Lung Injury

Jake Stroh introduced a data assimilation framework that models a wide variety of lung-ventilator system behaviors at MDS22.

New Surrogate Marker of Chronic Kidney Disease Progression and Mortality in Medical-word Virtual Space

Inherent knowledge in the medical literature is a candidate data source for the pathophysiology of chronic kidney...

Data Assimilation for the Prediction of Fault Slip Behavior in Subduction Zones

In pursuit of the ultimate goal to accurately predict earthquakes, a data assimilation system monitors slow fault...

Rare Event Sampling for the Earth and Planetary Sciences

A computational approach uses “survival of the fittest” to steer simulations toward the exhibition of extreme events.

Speeding Up Solar Wind Forecasts with Reduced-order Modeling

Shifted operator inference is a data-driven, projection-based, reduced-order model that predicts the solar wind...

Learning Oscillatory Navier-Stokes Flows and Causal Linear Operators with Deep Neural Network Algorithms

Neural networks address nonlinear problems with highly oscillatory solutions and approximate operators of causal...

Deep Learning for Digital Pathology: Detecting Skin Diseases with Normalizing Flows

Deep learning can assist in the diagnosis of skin diseases by defining changes in the skin as out-of-distribution samples.

Drift Matters: Unsupervised Anomaly Detection and National Security

Few existing algorithms can handle the concept drift challenges that are specific to maritime trajectories.

Predicting Hurricane Trajectories

Statistical modeling and machine learning methods vary in their effectiveness at predicting hurricane trajectories.

Download the Print Issue