Research interests of Uwe F. Mayer
For about the past two decades I switched my career and took on
research positions in industry. I am now what is known as a data
scientist and am working on large data sets employing statistical
learning, data mining and machine learning. I have several
publications and patents in these areas, supervised a PhD
student in machine learning at the University of California San Diego,
and I am a senior member of IEEE.
While working in academia, mathematically I was interested in partial
differential equations, evolution problems, free boundary problems,
global analysis, differential geometry, and numerical simulations. My
published mathematical work concentrates on geometric free boundary
problems. For those problems, one has an initial curve or surface and
a mathematical law which describes how the surface should evolve. The
best known example is the one where the normal velocity is the mean
curvature of the surface (possibly minus its average to force
preservation of enclosed volume). However, there are many others, for
example the Mullins-Sekerka flow, the surface diffusion flow, or the
Willmore flow. Follow this link if you want to see a few numerical simulations. Besides on free
boundary problems, I also previously worked on nonpositively curved
metric spaces, in particular on questions concerning gradients.
Research Publications
Clicking on a title will let you read an abstract and will give you
download options.
Data Mining / Machine Learning
-
Canary in the e-Commerce Coal Mine: Detecting and
Predicting Poor Experiences Using Buyer-to-Seller Messages (with D. Masterov
and S. Tadelis).
Proceedings of ec'15, Portland,
Oregon, USA (June 2015).
-
Bootstrapped Language Identification
For Multi-Site Internet Domains.
Proceedings of KDD'12, Beijing,
China (August 2012).
-
Classifying non-Gaussian and Mixed Data Sets
in their Natural Parameter Space
(with C. Levasseur and K. Kreutz-Delgado).
Proceedings of the Nineteenth IEEE
Int'l Workshop on Machine Learning for Signal Processing,
Grenoble, France (September 2009).
-
A Unifying Viewpoint of some Clustering
Techniques Using Bregman Divergences and Extensions to Mixed Data Sets
(with C. Levasseur, B. Burdge and K. Kreutz-Delgado).
Proceedings of the
1st International
Workshop on Data Mining and Artificial Intelligence (DMAI 2008) held
in conjunction with the 11th IEEE International Conference on
Computer and Information Technology, Khulna, Bangladesh (December 2008).
-
Generalized Statistical Methods for
Unsupervised Minority Class Detection in Mixed Data Sets
(with C. Levasseur, B. Burdge and K. Kreutz-Delgado).
Proceedings of the 2008
IAPR Workshop on Cognitive Information
Processing, Santorini, Greece, pp. 126-131 (2008).
-
Data-Pattern Discovery Methods for Detection
in Nongaussian High-dimensional Data Sets
(with C. Levasseur, K. Kreutz-Delgado, and G. Gancarz).
Conference Record of the
Thirty-Ninth
Asilomar Conference on Signals,
Systems and Computers, pp. 545-549 (2005).
-
Experimental Design for solicitation campaigns
(with A. Sarkissian). Proceedings of
KDD-2003, Washington,
DC, pp. 717-722 (2003).
Mathematics
-
Self-intersections for the Willmore flow
(with G. Simonett). Proceedings of the Seventh
International Conference on Evolution Equations: Applications to
Physics, Industry, Life Sciences and Economics - EVEQ2000. Nonlinear
Differential Equations Appl., 55, Birkhäuser, Basel, pp. 341--348 (2003).
-
A numerical scheme for axisymmetric
solutions of curvature driven free boundary problems, with
applications to the Willmore Flow (with G. Simonett).
Interfaces and Free Boundaries,
4, no. 1, pp. 89-109 (2002).
-
Numerical solutions for the surface
diffusion flow in three space dimensions.
Comput. Appl. Math.,
20, no. 3, pp. 361-379 (2001).
-
Loss of convexity for a modified
Mullins-Sekerka model arising in diblock copolymer melts (with
J. Escher).
Archiv
der Mathematik, 77, issue 5, pp. 434-448 (2001).
-
A singular example for the averaged mean curvature
flow. Experimental Math.,
10, no. 2, pp. 103-107 (2001).
-
A numerical scheme for free boundary problems
that are gradient flows for the area functional.
Europ. J. Appl. Math.,
11, issue 2, pp. 61-80 (2000).
-
Self-intersections for the surface diffusion
and the volume preserving mean curvature flow (with
G. Simonett). Differential
Integral Equations, 13, pp. 1189-1199 (2000).
-
On diffusion-induced grain-boundary motion
(with G. Simonett), in Nonlinear Partial Differential
Equations (Evanston, IL, 1998), Gui-Qiang Chen and Emmanuele
DiBenedetto (editors). Contemporary
Mathematics book series, Amer. Math. Soc., Providence, Rhode Island,
pp. 231-240, 1999.
-
Classical solutions for diffusion-induced
grain-boundary motion (with G. Simonett).
J. Math. Anal. Appl.,
234, pp. 660-674 (1999).
-
On the surface diffusion flow (with
J. Escher and G. Simonett), in Navier-Stokes equations and
related nonlinear problems (Palanga, 1997), VSP, Utrecht, pp. 69-79
(1998).
-
The surface diffusion flow for immersed
hypersurfaces (with J. Escher and G. Simonett).
SIAM
J. Math. Anal., 29, no. 6, pp. 1419-1433 (1998).
-
Gradient flows on nonpositively curved
metric spaces and harmonic maps.
Comm. Anal. Geom., 6,
no. 2, pp. 199-253 (1998).
-
Two-sided Mullins-Sekerka flow does not
preserve convexity, in Proceedings of the Third Mississippi
State Conference on Difference Equations and Computational Simulations
(Mississippi State, MS, 1997), J.Graef, R. Shivaji, B. Soni,
J. Zhu (editors), Electronic J.
Diff. Equ., Conference 01, 1997, pp. 171-179.
-
One-sided Mullins-Sekerka flow does not
preserve convexity.
Electronic J. Diff. Equ.,
1993 no. 08, pp. 1-7 (1993).
Graduate Students
-
Cécile Levasseur, Ph.D. (Electical Engineering), University of
California, San Diego (2009). Thesis:
Generalized Statistical Methods for Mixed Exponential Families,
jointly advised with Prof. Kenneth Kreutz-Delgado.
-
Carl-Heinz Kneisel, Diploma (Mathematics), Universität Hannover,
Germany (2003).
Thesis: On the global
existence of non-convex solutions for the averaged mean
curvature flow, advisor: Prof. Joachim Escher, numerical
simulations provided by Uwe F. Mayer.
Patents
-
Enhanced Supply And Demand Tool
(with W. Pang, J. Lane, J. Colmenares, G. Singh, R. Tiwari, P. Coles, D. Coey).
United States Patent Application 15/149,964 (Filed 2016/05/09, provisional application 62/159,052 filed 2015/05/08).
-
Fraud Detection System For The Faster Payments System
(with S. Zoldi).
United States Patent 11,037,229 (Granted 2021/06/15, application 20090222369 filed 2008/05/13).
-
Automated entity identification for efficient profiling in an event probability prediction system
(with A. Vaiciulis, L. Peranich, S. Zoldi and S. De Zilwa).
United States Patent 8,645,301 (Granted 2014/02/04, continuation application 13/710,423 filed 2012/12/10).
United States Patent 8,332,338 (Granted 2012/12/11, continuation application 13/399,067 filed 2012/02/17).
United States Patent 8,121,962 (Granted 2012/02/21, application 12/110,261 filed 2008/04/25).
-
Adaptive Fraud Detection
(with S. Zoldi, L. Peranich, J. Athwal and Sajama).
United States Patent 10,510,025 (Granted 2019/12/17, application 20090222243 filed 2008/02/29 titled Adaptive Analytics).
-
Fast Accurate Fuzzy Matching
(with V. Narayanan and M. Blume).
United States Patent 7,870,151 (Granted 2011/01/11, application 20080189279 filed 2007/06/13).
-
Comprehensive Identity Theft Protection System
(with T. Crooks and M. Lazarus).
United States Patent 8,296,250 (Granted 2012/10/23, continuation application 13/195,328 filed 2011/08/01).
United States Patent 7,991,716 (Granted 2011/08/02, continuation application 12/961,478 filed 2010/12/06).
United States Patent 7,849,029 (Granted 2010/12/07, application 11/421,896 filed 2006/06/02).
Other Publications
Contributed to Publications by Others
- Level Set Method: For Motion by Mean Curvature,
Notices of the AMS, November 2016, by Tobias Holck Colding and William P. Minicozzi II. (Uwe F. Mayer contributed some of the figures.)
-
Mean
Curvature Flow, Bulletin (New Series) of the American Mathematical Society 52 no. 2 (2015), by Tobias Holck Colding, William P. Minicozzi II, and Erik Kjær Pedersen. (Uwe F. Mayer contributed some of the figures.)
-
C++
by Dissection: The Essentials of C++ Programming, 1st Edition,
Addison Wesley Publishers, 2002, by Ira Pohl. (Uwe F. Mayer contributed some of the exercises.)
-
Carl-Heinz Kneisel, Diploma (Mathematics), Universität Hannover,
Germany (2003).
Thesis: On the global
existence of non-convex solutions for the averaged mean
curvature flow advisor: Prof. Joachim Escher. (Uwe F. Mayer contributed the numerical simulations.)
Acknowledged in Publications by Others
-
Optimizing Similar Item Recommendations in a
Semi-structured Marketplace to Maximize Conversion, 10th ACM
Conference on Recommender Systems, Boston, MA, USA, ISBN
978-1-4503-2138-9 DOI: 10.1145/1235 (2016), by Yuri M. Brovman,
Marie Jacob, Natraj Srinivasan, Stephen Neola, Daniel Galron, and
Ryan Snyder.
-
Linux Benchmarking:
Part 3 - Interpreting Benchmark Results, Linux Gazette, January 1998,
Issue 24, by André D. Balsa.
-
Linux Benchmarking:
Part 1 - Concepts, Linux Gazette, October 1997,
Issue 22, by André D. Balsa.
-
The proof of Fermat's Last Theorem by R. Taylor and
A. Wiles, written by Gerd Faltings, translated from German by Uwe F. Mayer.
Notices of the AMS,
42 no. 7, pp. 743-746 (1995).
-
Analysis 2,
1st Edition, Verlag Springer, 1995, by Wolfgang Walter.
Connectivity
-
Erdös Number: 5
Paul Erdös [0] ->
Svante Janson [1] ->
Jie Xiao1 [2] ->
Xuan Thinh Duong [3] ->
Gieri Simonett [4] ->
Uwe F. Mayer [5]
Paul Erdös [0] ->
Daniel Grieser [1] ->
Isamu Ohnishi [2] ->
Yasumasa Nishiura [3] ->
Joachim Escher [4] ->
Uwe F. Mayer [5]
Find your Erdös Number using the collaboration distance tool of MathSciNet.
-
Chomsky Number: 6
Noam Chomsky [0] ->
Marcel-Paul Schützenberger [1] ->
Alain Connes [2] ->
Don Bernard Zagier [3] ->
Jonatan Lenells [4] ->
Joachim Escher [5] ->
Uwe F. Mayer [6]