Cyber-physical systems are engineered systems that require tight conjoining of and coordination between the computational (discrete) and the physical (continuous). Cyber-physical systems are rapidly penetrating every aspect of our lives, with potential impact on sectors critical to national security and competitiveness, including aerospace, automotive, chemical production, civil infrastructure, energy, finance, healthcare, manufacturing, materials, and transportation. As these systems fulfill the promise of the Internet of Things, smart cities, household robots, and personalized medicine, we need to ensure they are trustworthy: reliable, secure, and privacy-preserving. This talk will look at cyber-physical systems from the lens of trustworthy computing. Throughout my talk, I will raise research challenges for how to make cyber-physical systems trustworthy, with a special emphasis on privacy.
Colloquium Series: Jeannette Wing
Colloquium Series: Dr. Robert Schapire
Principal Researcher, Microsoft Research (NYC)
Networks are increasingly important in many aspects of our world: physical networks like transportation networks, utility networks and the Internet, online information networks like the WWW, online social networks like Facebook and Twitter, epidemiological networks for global disease transmission, genomic and protein networks in computational biology, and many more. How do we model and learn these networks? In contrast to conventional learning problems, where we have many independent samples, it is often the case for these networks that we can get only one independent sample.
Director of Basketball Analytics, NBA
Michael Lewis's 2003 bestseller Moneyball illustrated the power of effectively utilizing analytics in baseball. His book ignited a sports analytics revolution by showing how data-driven decision making processes could be as valuable as a multimillion dollar advantage in player spending.
In the past several years, an online basketball analytics community (APBR) has flourished, developing tools to evaluate teams, players and strategies. Basketball teams have been hiring analysts at an ever-increasing rate, and even the style of NBA play is changing before our eyes.
Furthermore, starting in 2013, every NBA arena became equipped with the SportVU software. Utilizing six cameras installed in the catwalks of every NBA arena, SportVU tracks the movement of every player on the court and the basketball 25 times per second.
Dr. David Heckerman
Distinguished Scientist, Microsoft Research
One obvious way to get more power out of association studies is to analyze more data. Large studies, however, are often confounded by relatedness among individuals, leading to the need for sophisticated statistical methods such as the linear mixed model. Unfortunately, these models have traditionally required an inordinate amount of computation.
Dr. Heckerman will show how these methods can be made very fast, facilitating their use on large data and will also describe some additional tricks to get more power out of association studies.
Dr. Yoram Singer reviews the design, analysis, and implementation of stochastic optimization techniques, online algorithms, and modeling approaches for learning in high dimensional spaces using large amounts of data. His focus is on algorithms and models that are efficient, accurate, and yield compact models. Concretely, his group describes the forward-backward shrinkage algorithm (Fobos), mirror descent for learning composite objectives (COMID), and the stonking adaptive gradient (AdaGrad) algorithm.
A Survey of Probabilistic Programming
Probabilistic inference is a widely-used, rigorous approach for processing ambiguous information based on models that are uncertain or incomplete. However, models and inference algorithms can be difficult to specify and implement, let alone design, validate, or optimize. Additionally, inference often appears to be intractable. Probabilistic programming is an emerging field that aims to address these challenges by formalizing modeling and inference using key ideas from probability theory, programming languages, and Turing-universal computation.
Data science holds the promise to solve many of society's most pressing challenges but much of the necessary data is locked within the volumes of text and speech on the web. Thus, in many cases, data science can only succeed if paired with natural language processing. In this talk, Professor McKeown will describe research projects that draw from language data along a continuum from fact to fiction.