A Conversation with Grace Wahba

Douglas Nychka, Ping Ma, Douglas M. BatesStatistical 2020 · DOI

Abstract

Grace Wahba (née Goldsmith, born August 3, 1934), I. J. Schoenberg-Hilldale Professor of Statistics at the University of Wisconsin-Madison (Emerita), is a pioneer in methods for smoothing noisy data. Her research combines theoretical analysis, computation and methodology motivated by innovative scientific applications. Best known for the development of generalized cross-validation (GCV), the connection between splines and Bayesian posterior estimates, and “Wahba’s problem,” she has developed methods with applications in demographic studies, machine learning, DNA microarrays, risk modeling, medical imaging and climate prediction. Grace grew up in the Washington, DC area and New Jersey, and graduated from Montclair High School. She was educated at Cornell (B.A. 1956), University of Maryland, College Park (M.A. 1962) and Stanford (Ph.D. 1966), and worked in industry for several years before receiving her doctorate in 1966 and settling in Madison in 1967. Although holding several visiting appointments, she has made Madison her home for over 50 years. She is the author of Spline Models for Observational Data which has garnered more than 8000 citations. Grace is treasured as an academic advisor and has mentored 39 Ph.D. students that have resulted in more than 330 academic descendants. She was elected to the United States National Academy of Sciences in 2000 and received an honorary degree of Doctor of Science from the University of Chicago in 2007. Wahba is a Fellow of several academic societies including the American Academy of Arts and Sciences, the American Association for the Advancement of Science, the American Statistical Association and the Institute of Mathematical Statistics. Over the years, she has received a selection of notable awards in the statistics community: R. A. Fisher Lectureship (2014), Gottfried E. Noether Senior Researcher Award (2009), Committee of Presidents of Statistical Societies Elizabeth Scott Award (1996) and the first Emanuel and Carol Parzen Prize for Statistical Innovation (1994).

Introduction

Grace Wahba is a foundational figure in modern statistics, best known for pioneering Generalized Cross-Validation (GCV) and the Reproducing Kernel Hilbert Space (RKHS) framework for splines. Her career, documented in the biographical interview "A Conversation with Grace Wahba" (Statistical Science, 2020), traces a remarkable journey from working as a "human calculator" to becoming a titan of the field whose work underpins much of modern machine learning and smoothing methodology.

Beyond her mathematical achievements, Wahba's career highlights the historical challenges faced by women in STEM. Navigating her path as a single mother in the 1950s and 60s, she worked in environments where women were often excluded from professional social circles or faced systemic discrimination. Her success in this context adds a layer of resilience to her intellectual legacy.

Theoretical Foundations and Methodology

Wahba’s work is characterized by the unification of rigorous theoretical analysis with computational practicality. Several landmark methods emerged from her research:

  • The Kimeldorf-Wahba Representer Theorem: This fundamental result demonstrates that the solution to an optimization problem in an infinite-dimensional RKHS can be expressed as a finite linear combination of kernel functions evaluated at the data points. This theorem is a cornerstone of kernel methods.
  • Generalized Cross-Validation (GCV): Developed to address the challenge of choosing the optimal smoothing parameter in splines, GCV emerged from the realization that standard leave-one-out cross-validation was computationally prohibitive and lacked rotation invariance. By utilizing the trace of the influence matrix, Wahba and her collaborators created a robust and efficient selection criterion.
  • RKHS Framework for SVMs: Wahba was instrumental in bridging the gap between statistics and machine learning by showing that Support Vector Machines (SVMs) could be viewed as regularization problems in an RKHS. By replacing the squared error loss of splines with a "hinge loss" function, she brought SVMs into the statistical mainstream.
Announcement for the COPSS Award and Fisher Lecture, highlighting Wahba's work on RKHS.

Scientific Applications and "Wahba’s Problem"

Wahba’s theoretical breakthroughs were frequently motivated by diverse, real-world scientific problems. One of her most famous contributions, "Wahba’s Problem," arose from satellite attitude determination. She developed a least-squares method to find the optimal rotation matrix required to reconcile observed star positions with known catalogs. Interestingly, she only discovered the problem was named after her 15 years after publishing the solution—a classic example of Stigler’s Law of Eponymy.

Her work also extended to meteorology, using spatial observations to refine numerical weather prediction models, and biomedicine, where she applied splines and SVMs to DNA microarrays and tumor size distributions.

Legacy and Recognition

Grace Wahba’s impact is measured not only in citations—her seminal book Spline Models for Observational Data has over 8,000—but also in her extensive academic lineage. Having mentored 39 Ph.D. students, she has over 330 academic descendants, many of whom continue to push the boundaries of the field.

Grace Wahba receiving an honorary Doctor of Science from the University of Chicago in 2007.

Her recognition includes election to the National Academy of Sciences and numerous prestigious awards, such as the R.A. Fisher Lectureship. Throughout her career, she maintained that statistics must remain at the center of the "scientific enterprise," advocating for deep collaboration with subject-matter experts to ensure that theoretical rigor meets practical necessity.

Summary updated

Abstract

Grace Wahba (née Goldsmith, born August 3, 1934), I. J. Schoenberg-Hilldale Professor of Statistics at the University of Wisconsin-Madison (Emerita), is a pioneer in methods for smoothing noisy data. Her research combines theoretical analysis, computation and methodology motivated by innovative scientific applications. Best known for the development of generalized cross-validation (GCV), the connection between splines and Bayesian posterior estimates, and “Wahba’s problem,” she has developed methods with applications in demographic studies, machine learning, DNA microarrays, risk modeling, medical imaging and climate prediction. Grace grew up in the Washington, DC area and New Jersey, and graduated from Montclair High School. She was educated at Cornell (B.A. 1956), University of Maryland, College Park (M.A. 1962) and Stanford (Ph.D. 1966), and worked in industry for several years before receiving her doctorate in 1966 and settling in Madison in 1967. Although holding several visiting appointments, she has made Madison her home for over 50 years. She is the author of Spline Models for Observational Data which has garnered more than 8000 citations. Grace is treasured as an academic advisor and has mentored 39 Ph.D. students that have resulted in more than 330 academic descendants. She was elected to the United States National Academy of Sciences in 2000 and received an honorary degree of Doctor of Science from the University of Chicago in 2007. Wahba is a Fellow of several academic societies including the American Academy of Arts and Sciences, the American Association for the Advancement of Science, the American Statistical Association and the Institute of Mathematical Statistics. Over the years, she has received a selection of notable awards in the statistics community: R. A. Fisher Lectureship (2014), Gottfried E. Noether Senior Researcher Award (2009), Committee of Presidents of Statistical Societies Elizabeth Scott Award (1996) and the first Emanuel and Carol Parzen Prize for Statistical Innovation (1994).
BibTeX
@inproceedings{nychka2020conversation,
  author = {Douglas Nychka and Ping Ma and Douglas M. Bates},
  title = {{A Conversation with Grace Wahba}},
  booktitle = {Statistical},
  year = {2020},
  doi = {10.1214/19-sts734},
}
Figure