Research Projects Awarded Funding Through Partnership with American Family Insurance

Nine UW–Madison data science research projects with applications to both the insurance industry and society have been selected for funding through the American Family Funding Initiative.

The UW–Madison Data Science Institute (DSI), powered by American Family Insurance, announced these awards for research topics ranging from assessing wildfire risk to reducing the computational costs of extreme weather prediction and auditing large language models for bias and fairness.

Through a unique sponsored research partnership, American Family Insurance has committed $10 million over 10 years for UW–Madison research with potential to fuel discovery in data science. Since its launch in spring 2020, 40 teams of UW–Madison faculty and collaborators have been awarded nearly six million dollars through this initiative.

“We’re thrilled to announce that, in our latest round of collaboration, we’ve really brought together some amazing minds from UW-Madison and our own teams at American Family Insurance,” said Zak Rottier, Enterprise Director of Data Science at American Family Insurance. “It’s all about connecting the dots between the brilliant researchers and our strategic insurance goals. Year after year, this partnership continues opening up new avenues for innovation that benefit those directly involved, as well as our American Family Insurance customers and the broader data science community.”

Funds are awarded through a competitive application process administered by DSI, and American Family Insurance provides selected projects with both funding and collaboration with their data scientists. Both organizations evaluate applications on their novelty, potential impact to data science and alignment with topics of interest to American Family.

“The projects awarded in round 6 highlight the diversity of challenges that industries face and the multitude of ways that data science techniques can address those challenges. The creative solutions that are proposed run the gamut from generative AI to more traditional approaches from statistics,” said Kyle Cranmer, David R. Anderson Director of the Data Science Institute.

A seventh round of funding will be announced in January 2025. Project summaries and PI bios are available at the data science @ uw website.

American Family Funding Initiative Round 6 Awards:

Multimodal Method Approach for Risk Assessment and Reasoning
PI: Kaiping Chen, Assistant Professor, Department of Life Sciences Communication
Co-PI: Junjie Hu, Assistant Professor, Departments of Computer Sciences and Biostatistics and Medical Informatics

Data-Driven Wildfire Ignition Prediction from an Insurance Perspective
PI: Min Chen, Assistant Professor, Department of Forest and Wildlife Ecology

Multimodal Foundation Model for Driver Behavior Profiling and Risk Analysis
PI: Song Gao, Associate Professor, Department of Geography

Sensitivity Analysis and Counterfactual Analysis of Insurance Data
PI: Hyunseung Kang, Associate Professor, Department of Statistics

A Framework for Valid and Reliable Audits of Biases in Large Language Models
PI: Shamya Karumbaiah, Assistant Professor, Department of Educational Psychology
Co-PI: Daniel Bolt, Professor, Department of Educational Psychology

Interpretable Causal Inference for Multimodal and Relational Data
PI: Keith Levin, Assistant Professor, Department of Statistics

Safer Driving Through Optimized Telematics-Based Feedback
PI: Tony McDonald, Assistant Professor, Department of Industrial and Systems Engineering
Co-PI: Yonatan Mintz, Assistant Professor, Department of Industrial and Systems Engineering

Novel Methods for Hail Detection
PI: Kevin Ponto, Associate Professor, Wisconsin Institute for Discovery and School of Human Ecology
Co-PI: Ross Tredinnick, Systems Programmer, Wisconsin Institute for Discovery

Developing a Prototype Data-Driven Stochastic Convective Hazards Emulator
PI: Daniel Wright, Professor, Department of Civil and Environmental Engineering
Co-PI: Yagmur Derin, Research Scientist, Department of Civil and Environmental Engineering