Contact Us

Give us a call or drop by anytime, we endeavor to answer all inquiries within 24 hours.


Find us

PO Box 16122 Collins Street West Victoria, Australia

Email us /

Phone support

Phone: + (066) 0760 0260 / + (057) 0760 0560

Loading Events

« All Events

  • This event has passed.

Data Science Insights Speaker Series: Yian Ma

June 23, 2021 @ 5:30 pm - 7:00 pm


Exploration Problem in Sequential Decision Making: A Computational Perspective
by Yian Ma

Efficient Exploration is often the bottleneck for solving sequential decision making problems. Many different approaches have been proposed and analyzed, such as explore-then-commit, upper confidence bound, etc. Much of the focus has been on using frequentist perspectives to understand and develop entirely model-free or model-based methods. In practice, we often have some information about the system and can benefit from a generative model that has the flexibility of incorporating new information at different stages of the learning process.

In this talk, Yian will discuss how to design scalable computational methods that learn from the generative model and ensure that the optimal regret is achieved with a constant computational budget. That requires us to have increasingly accurate estimation with a growing data set, under a constant number of iterations and computation per iteration. He will present a stochastic gradient Markov chain Monte Carlo algorithm to achieve this goal.

Yian Ma is an assistant professor at the Halıcıoğlu Data Science Institute and an affiliated faculty member at the Computer Science and Engineering Department of the University of California San Diego. Prior to UCSD, he spent a year as a visiting faculty at Google Research. Before that, he was a post-doctoral fellow at EECS, UC Berkeley. He completed his Ph.D. at the University of Washington. His current research primarily involves scalable inference methods and their theoretical guarantees. He has been designing new Bayesian inference algorithms (with a focus on applying them to time series data and sequential decision making) that are provably efficient in terms of computational and statistical guarantees.

Agenda (Pacific Daylight Time, UTC -07)
– 5:30 – 5:40 pm — Gathering and introductions
– 5:40 – 6:30 pm — Talk
– 6:30 – 7:00 pm — Q & A, discussion

Links to slides and videos of meetup presentations are available on the SDML GitHub repo

Join our slack channel or leave a comment below if you have any questions about the group or need clarification on anything.


June 23, 2021
5:30 pm - 7:00 pm
Event Categories: