Aarhus University Seal / Aarhus Universitets segl

Bo Thiesson

Adjungeret professor, Timelærer

Education

1996
Ph.D. in Statistics, Aalborg University, Denmark
Dissertation: “Some Statistical Methods for Learning Probabilistic Networks”

1991
M.Sc. in Computer Science, Aalborg University, Denmark
Thesis: “(G)EM Algorithms for Recursive Graphical Association Models with Missing Data”

1989
B.Sc. in Computer Science, Aalborg University, Denmark

1987
B.Sc. in Mathematics, Aalborg University, Denmark

1992-93
Visiting Scholar, Department of Statistics, Stanford University, USA


Career

2016 - present
Honorary Professor, Department of Engineering, Århus University, Denmark

2015 - present
Information Engineer, CTO, ENVERSION A/S, Denmark

2013 - present
Associate Professor, Department of Computer Science, Aalborg University, Denmark (partial leave since 2015)

2000-2012
Researcher, Machine Learning & Applied Statistics Group, Microsoft Research, Redmond, WA, USA

1999-2000
Research Software Design Engineer, Decision Theory & Applied Statistics Group, Microsoft Research, Redmond, WA, USA

1996-1999
Software Design Engineer, Decision Theory & Applied Statistics Group, Microsoft Research, Redmond, WA, USA

1995-1996
Scientist, DINA Aalborg, Department of Mathematics and Computer Science, Aalborg University, Denmark

1991-1995
Research Assistant, Department of Mathematics and Computer Science, Aalborg University, Denmark



Publications - Summary

-h-index: 18-41 publications (35 peer-reviewed); predominantly in 1-tier journals and conferences (JMLR, Machine Learning, Bayesian Statistics, KDD, UAI, NIPS, AI&Statistics, ECCV, SIGIR, UIST, HCI).-Publications have been cited in total 1421 times. (On average 68 cites/year.)



Patents - Summary

-38 US Patents awarded.
-7 US Patent applications under evaluation by the patent office.


Teaching Experience – Summary

1992, 1994, 1996
Lecturer – “Graphical Models”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Mathematics and Computer Science.
Developed syllabus and overall course structure.

1993
Lecturer – “Recursive Models & the EM Algorithm”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Mathematics and Computer Science.
Developed syllabus and overall course structure.

2013
Lecturer – “Machine Learning”, Aalborg University, Denmark
PhD course at the Doctoral School of Engineering and ScienceResponsible for “Big-Data” part of this course.

2013, 2014, 2015
Lecturer – “Advanced Topics in Machine Intelligence”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Computer Science.
Co-developed syllabus and overall course structure.

2013, 2014, 2015
Lecturer – “Advanced Topics in Machine Intelligence”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Computer Science.
Co-developed syllabus and overall course structure.

2013, 2014
Lecturer – “Specialization course in Machine Intelligence”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Computer Science.
Co-developed syllabus and overall course structure.

2014, 2015
Lecturer – “Web Intelligence”, Aalborg University, Denmark
M.Sc. graduate course at the Dept. of Computer Science.
Developed syllabus and overall course structure.

2004-2012
Manager & mentor for 10 Ph.D. Internships. Microsoft Research, Redmond, WA, USA
Defined and supervised 12 weeks long internship projects, giving the opportunity for outstanding Ph.D. students to apply research & theory to practice at Microsoft Research.

2013 - present
Project supervisor for a very large number of project groups and 11 Master’s Thesis students, Aalborg University, Denmark.
I am currently supervising 12 Master’s Thesis students towards their graduation next year.

2013 - present
Co-supervisor for 2 Ph.D. students, Microsoft Research, Redmond, WA, USA.(With the focus on healthcare in my new position at ENVERSION A/S, it has been necessary for me to discontinue formal supervision of these PhD students and I am now only helping them on a friendly basis.)

2013
Participated in “Adjunktpædagogikum” (University Pedagogy) by AAU Learning Lab. Aalborg University, Denmark.
Passed at highest possible evaluation level.



Awards

2001
Gold Star Award*: For work on large-scale learning for mixtures of Bayesian networks.  *Microsoft Award Programme: “The Gold Star Award is highly selective. It is given in recognition of important accomplishments that contribute to Microsoft’s business success.”

2006
Gold Star Award*: For work on handwriting recognition & user personalization.

2007
Winner adPrediction Competition: Internal Microsoft competition setup to advance Microsoft AdCenter’s Relevance and Revenue algorithms (we were two groups sharing the honor).

2001-2011
Ship-It awards: For technology transfers to Windows-XP, Microsoft Commerce Server 2002, SQL Server 2005, Windows Vista, Microsoft Excel 2007, Microsoft AdCenter Keyword Services Platform, Windows Phone 6, 6.5, & 7, Microsoft BI Labs (SQL Server Data Mining in the Cloud), and Microsoft AdCenter Capacity Prediction Services (internal tool).



Product Impact (At Microsoft)

Data mining for SQL Server: Contributed to the data mining tools in SQL Server 2005 and beyond. Also included in the SQL Server 2008 Data Mining Add-ins for Office and this work is getting a second life in Microsoft BI Labs (SQL Server Data Mining in the Cloud). 30% of people who use SQL Analysis Services use data mining; ~22k members on sqlserverdatamining.com.

WildThing -“fast regexp matching”: As part of the AutoSuggestEngine (ASE) it has shipped in the Bing Mobile client for Windows Mobile 6.x, shipped in the Bing Mobile Blackberry, and Brew clients, shipped in the Mobile Text Framework of Windows Phone 7. It has also shipped as the “FastContainedKeyword” functionality in the AdCenter Keword Service Platform V3 and beyond.

TabletPC, handwriting recognition: Personalization, URL recognizer, and contributed to multi-stroke Asian recognition. The personalization was one of the most prominent features mentioned in C|Net article: “Windows Vista offers handwriting options”.

Time Series Foundation –TSF: We shipped TSF with AdCenter, Capacity Prediction Services, Jan. 2011. TSF is also featured (internally & externally) on the Microsoft BI Labs portal, where is has had almost 1K downloads (in less than a year).

Excel Color Scales: Contributed to the research and demonstration of an abnormality detector that inspired Excel 2007's Color Scales. http://msdn2.microsoft.com/en-us/library/bb428945(office.11).aspx

Many, many consulting roles across a large variety of Microsoft product groups.



Academic Services over the Years

Senior PC: UAI

PC: UAI, ICML, KDD, AI&Statistics, ICPRAM, ECML, ACL

Journal reviewer: Bayesian Analysis; Physica A; IEEE Transactions on Pattern Analysis and Machine Intelligence; IEEE Transactions on Signal Processing; Journal of the American Statistical Association; Computational Statistics & Data Analysis; Journal of Machine Learning Research.