FIELDS INSTITUTE FOR RESEARCH IN MATHEMATICAL SCIENCES
April 7, 2015 at 3:30 p.m.
April 8, 2015 at 3:30 p.m.
April 9, 2015 at 11:00 a.m.
Fields Institute, Room 230
of California at Berkeley
7, 3:30 p.m.
On Computational Thinking, Inferential Thinking and "Big Data"
rapid growth in the size and scope of datasets in science and technology has
created a need for novel foundational perspectives on data analysis that blend
the inferential and computational sciences. That classical perspectives from
these fields are not adequate to address emerging problems in "Big Data"
is apparent from their sharply divergent nature at an elementary level---in
computer science, the growth of the number of data points is a source of "complexity"
that must be tamed via algorithms or hardware, whereas in statistics, the
growth of the number of data points is a source of "simplicity"
in that inferences are generally stronger and asymptotic results can be invoked.
I present several research vignettes on topics at the computation/statistics
interface, including the problem of trading off inference and privacy, the
problem of inference under communication constraints and algorithmic weakening
as a tool for trading off the speed and accuracy of inference.
8, 3:30 p.m.
Lower Bounds at the Computational and Statistical Interface
of the grand challenges of our era is the attempt to bring computational
and statistical ideas together in a theoretically-grounded framework for
scalable statistical inference. This is made challenging by the lack of
a role for computational concepts such as "runtime" in core statistical
theory and the lack of a role for statistical concepts such as "risk"
in core computational theory. I discuss further attempts to build bridges
between "computational thinking" and "inferential thinking,"
focusing on the theoretical study of lower bounds that embody computational
and statistical constraints.
April 9, 11:00 a.m.
Distributed Computing, the Bootstrap and Concurrency Control
Divide-and-conquer is a powerful paradigm in computer science, informing
the design of algorithms and the design of distributed computing architectures.
In the statistical setting, care must be taken, because naive use of divide-and-conquer
can yield incorrect statistical inferences. Focusing on distributed computing
architectures, I present a distributed version of the bootstrap and discuss
the use of concurrency control to trade off the speed and accuracy of inference.
Michael I. Jordan is the Pehong Chen Distinguished Professor in the Department
of Electrical Engineering and Computer Science and the Department of Statistics
at the University of California, Berkeley. His research interests bridge the
computational, statistical, cognitive and biological sciences, and have focused
in recent years on Bayesian nonparametric analysis, probabilistic graphical
models, spectral methods, kernel machines and applications to problems in
distributed computing systems, natural language processing, signal processing
and statistical genetics. Prof. Jordan is a member of the National Academy
of Sciences, a member of the National Academy of Engineering and a member
of the American Academy of Arts and Sciences. He is a fellow of the American
Association for the Advancement of Science. He has been named a Neyman Lecturer
and a Medallion Lecturer by the Institute of Mathematical Statistics. He received
the David E. Rumelhart Prize in 2015 and the ACM/AAAI Allen Newell Award in