Tomas Singliar

I am a Senior Data Scientist at Microsoft.

We deploy Azure-based reference solutions for data science problems; my team specializes in attaching predictive insights to Microsoft Dynamics implementations.

If machine learning is your thing, you should check out Azure ML Studio. It really is awesome - it still blows me away that there is a single button to deploy your model as a web service once you develop it, and it works!

My newest research interests tend to be driven by business need. Recently:

  • Causality, of the Pearl sort, to do sound causal inferences from observational data
  • Econometrics, causality of the Rubin sort
  • Experiment (A/B testing) design and analysis

My older but still active interests include:

  • Machine Learning, especially graphical probability models
  • Action understanding through IRL
  • Data collection and mining, sensor data analytics at terabyte+ scales

Contact information

  • (+1) 425 722 8627 (business, email me for cell)
  • work: Tomas.Singliar@(my company).com
  • personal: last name at google's public mail service

My Research and Development at Amazon

Causality from non-experimental data (proprietary)

My research is mostly proprietary. In general terms, I estimate the causal effect of making improvements to the catalog data to the key perfomance metrics such as page views and purchases. This is extraordinarily complex because of the number of subsystems that each affect customer experience, and feed off the catalog data. It is not classical A/B testing for reasons of scalability, we have to make do with observational data. That has all sorts of statistical validity issues that we need to mitigate. We built and maintain a system using a large scale data pipeline feeding an econometric model that does the estimation.

Statistical consulting

Design and analyze experiments to answer business questions, define sampling protocols to deal with wild data exhibiting power-laws distributions, ...

My Research and Development at Boeing

Before Amazon, I was an Advanced Technologist (research scientist in disguise) at Boeing Research and Technology in Bellevue, WA. These are the things I've done there. Due to proprietary nature of the projects, the descriptions are deliberately vague - sorry.

Large-scale sensor data analytics (proprietary)

How does one build self-service predictive data analytics for engineers who are not experts in computing, but rather the system where the data originates?

The XDATA project

Program website.

Building a highly-scalable Bayesian network library based on SMILE with University of Pittsburgh's DSL folks.

Understanding Purposeful Behavior

Using methods of inverse reinforcement learning, computers actively learn from humas to really understand observed behavior (defined as: ascertain and interpret the incentives and beliefs that explain the behavior as rational) of large numbers of agents, creating a ISR data exploitation capability to concentrate analaysts' attention on unusual and suspicious behavior instances, alert and generate explanations of the observations.

Publications:

DARPA Bootstrapped Learning (Phases 2 and 3)

The BL program attempts to implement the "Bootstrap Learning Dream", which is to dispense with the need for programming. Instead, the agent (such as a UAV) is taught how to perform the required tasks by somebody who understands the problem, instead of understanding programming.

Agent Executive based on Partial-Order Planning (proprietary)

Wrote a simple executive for an autonomous agent that uses the "repairability" of partial order plans to react in an environment where actions are nearly deterministic (so you can ignore uncertainty in planning), but robust recovery from action or resource failure is essential.

Automatic Derivation of Decision Policies (proprietary)

How to do reinforcement learning when rewards are large but rare. Automating reward shaping.

Computer Vision & Machine Learning applied to Satellite Imagery (proprietary)