Tomas Singliar
Curriculum vitae
http://www.singliar.info
singliar@gmail.com
412-251-1966
Education



2008 PhD in Computer Science, University of Pittsburgh. Applied Machine Learning.
2005 MS in Computer Science, University of Pittsburgh. Statistical Machine Learning.
2001 MSc (“magister”) in Informatics, Comenius University, Bratislava, Slovakia. Formal
methods, software and systems engineering.
Skills
Soft and Managerial Skills


Expert: Personnel Development (hiring, technical mentoring, performance management),
KPI and Business Metrics
Skilled: Communication and Writing
Hard Technical Skills


Expert: Machine Learning, Data Science, Predictive Analytics
Skilled: Optimization (linear and convex programming), Software Development, Big Data
Systems and Tools, Business Intelligence
Tools (hands-on experience)



Machine Learning: R, Matlab, SMILE (Bayes Net library), CPLEX
Software Development: Java, C++, C#, SQL – varying OS and platforms
Big Data: Amazon Redshift, the Hadoop stack
Current employment
(Sep 2013 – present) Research Scientist III, Search and Discovery, Amazon.com, Seattle, WA





Design and deploy econometric and predictive models to estimate causal impact of improving
Amazon catalog on key performance metrics (page views, sales) from observational data.
Provide technical guidance within and beyond the team of 50 SDEs on sampling, statistics,
and advanced algorithms
Influence definition of key annual organizational goals and metrics, supervise their
implementation, propose and review course corrections
Publish internally and disseminate knowledge and practices across organization boundaries
Please see my web page for (necessarily vague) project information
Previous professional experience and membership





(Feb 2009 – Aug 2013) Advanced Technologist 4, Boeing, Seattle, WA
o Develop innovative cutting-edge machine learning and data mining technology that
Boeing can license and/or turn into new products and services
o Understand Boeing business and technology challenges, design, implement and deliver
applicable machine learning or data-driven cost-saving solutions to business units
o Acquire and lead contracted research and development, mainly US government
o Please see my web page for (necessarily vague) project information
(Fall 2008) Business Intelligence Intern, Guru.com. Developed a model predicting the winner
of project bids in an online marketplace.
(Summers 2006, 2007) Graduate Technical Intern, Intel Corporation. Distributed detection of
network intrusions with probabilistic graphical models.
(Summer 2005) Dimension 5, Slovakia. Developed component analysis and fast k-means
modules for an advanced data visualization software package.
Other minor contracts in systems administration, software development and business analytics
Patents




T. Singliar, W. Murray, R. Cranfill, D. Margineantu: Natural language interface for systems
reasoning about observed behavior, submitted 2013, patent pending
T. Singliar, D. Margineantu: Intent estimation method and system for agents of limited
perception. US Patent #8,959,042. Amends USP#8,756,177.
T. Singliar, D. Margineantu: Methods and system for estimating subject intent from
surveillance, US Patent #8,756,177, issued June 2014
T. Singliar: Monitoring the state-of-health information for components, US Patent #8,533,133,
issued Sep 2013
Peer-reviewed publications
In Journals


T. Singliar, M. Hauskrecht: Noisy-OR Component Analysis and its Application to Link
Analysis. Journal of Machine Learning Research, vol. 7, October 2006
T. Singliar, M. Hauskrecht: Learning to detect incidents from noisily labeled data, Machine
Learning Journal, vol. 79, pages 335-354, September 2010
In Conference Proceedings



T. Singliar, F. Moerchen: DELi: A framework for measuring customer impact of catalog
changes, Amazon Machine Learning Conference AMLC’2015 [internal publication]
N. Rose, A. Dutta, T. Singliar: A quasi-A/B technique for ASIN experiments, Amazon
Machine Learning Conference AMLC’2015 [internal publication]
T. Singliar, F. Moerchen: Quantifying impact of sourcing catalog data, Amazon Machine
Learning Conference AMLC’2014 [internal publication]








T. Singliar, D. Dash: Efficient Inference in Persistent Dynamic Bayesian Networks; 24th
Conference on Uncertainty in Artificial Intelligence, 2008
T. Singliar, M. Hauskrecht: Approximation Strategies for Routing in Dynamic Stochastic
Networks; 10th International Symposium on Artificial Intelligence and Mathematics, 2008
T. Singliar, M. Hauskrecht: Learning to Detect Adverse Traffic Events from Noisily Labeled
Data; 11th European Conference on Principles and Practice of Knowledge Discovery in
Databases, 2007
T. Singliar, M. Hauskrecht: Modeling Highway Traffic Volumes; 18th European Conference
on Machine Learning, 2007
T. Singliar, D. Dash: COD: Online Temporal Clustering for Outbreak Detection; 22nd
Conference on Artificial Intelligence, 2007
T. Singliar, M. Hauskrecht: Variational Learning for Noisy-OR Component Analysis; SIAM
Conference on Statistical Data Mining, 2005
M. Hauskrecht, T. Singliar: Monte-Carlo optimization for resource allocation problems in
stochastic network systems; International Conference on Uncertainty in Artificial Intelligence,
2003
G. Juhas, R. Lorenz, T. Singliar: On synchronicity and concurrency in Petri Nets; Proceedings
of Applications and Theory of Petri Nets, 2003
In Workshop Proceedings



T. Singliar, D. Margineantu: Scaling up Inverse Reinforcement Learning through Instructed
Feature Construction, The “Snowbird” Learning Workshop, 2011
T. Singliar, M. Hauskrecht: Towards a Learning Incident Detection System. Workshop on
Machine Learning for Surveillance and Event Detection, International Conference on Machine
Learning 2006
T. Singliar, M. Hauskrecht: Modeling Large Stochastic Networks. Workshop on Robust
Communication in Complex Networks; Neural Information Processing Systems 2003
Service on conference and journal editorial committees






Senior Program Committee Member, AAAI-10 and -12, KDD 2015
Program committee member/Reviewer for all major conferences in the AI and ML domains.
In last 3 years: AAAI, IJCAI, UAI, ECML/PKDD, KDD (most every year)
Co-Chair (local arrangements) for ICML 2011 (28th International Conference on Machine
Learning)
Reviewer, Machine Learning Journal, 2010
Reviewer, Journal of Pattern Recognition Research, 2009
Reviewer, Journal of Applied Statistics, 2009
Research experience

2013-present Research Scientist, Amazon.com.




2009-2013 Advanced Technologist, Boeing Research and Technology, multiple machine
learning and artificial intelligence designing future intelligent technologies
2007 Graduate Technical Intern, Intel Research. Developed a much faster inference algorithm
for a subclass of dynamic Bayesian networks, with Dr. Denver Dash.
2007 Graduate Technical Intern, Intel Research. Developed a new technique for detection of
day-zero network worm intrusions that resulted in 40% reduction of worm penetration at time
of detection, with Dr. Denver Dash.
2002-2008 Graduate Student Researcher, University of Pittsburgh. Design and implementation
of novel data processing and machine learning algorithms in Matlab and C++ to support
research goals, with Dr. Milos Hauskrecht
Teaching and mentorship



2014 Amazon.com Intern Mentor (SDE and Business)
2012 Volunteer Mentor for the National Association of Black Engineers
2002-2005 Teaching assistantships, University of Pittsburgh, 4 computer science courses
Personal




Citizen of Slovak Republic (authorized for employment EU-wide)
Permanent Resident of the USA
Languages: Slovak/Czech native, English native level, French basic
Interests: mountaineering, astronomy, books.