Skip to main content

Regular Seminars

Welcome to CRiSM seminar series!

Seminars take place approximately biweekly in term time, in the Department of Statistics. There will be wine and cheese after the talks in the Statistics Common Room (C0.06).

We encourage all postgraduate students (MSc and PhD) to attend this series: it is a great opportunity to know more about current research within the department and outside.

CRiSM seminars 2017/2018 are organised by Dr Shahin Tavakoli.


Show selected | Select all / none
  • APTS
  • ASRU
  • Admissions
  • CRiSM
  • CRiSM Seminars
  • CRiSM Workshops
  • Dept
  • IT Committee
  • MASDOC
  • MG Group
  • MPTS
  • Master Class
  • Open Day
  • OxWaSP
  • P@W
  • P@W Workshops
  • Postgraduate
  • Public Engagement Seminar
  • RSS
  • Reading Group
  • Research Committee
  • Research Group
  • SF@W
  • SSLC
  • Seminars
  • Teaching Committee
  • Undergraduate
  • WCC
  • WDSI
  • WDSI Workshops
  • Workshops
  • YRM
  • i-like
  • Untagged
 
Help using calendars
 
Fri, Oct 27, '17
15:00 - 16:00
CRiSM Seminar
A1.01
3-4pm A1.01, Oct 27, 2017 - Azadeh Khaleghi

Title: Approximations of the Restless Bandit Problem

Abstract: In this talk I will discuss our recent paper on the multi-armed restless bandit problem. My focus will be on an instance of the bandit problem where the pay-off distributions are stationary ϕ-mixing. This version of the problem provides a more realistic model for most real-world applications, but cannot be optimally solved in practice since it is known to be PSPACE-hard. The objective is to characterize a sub-class of the problem where good approximate solutions can be found using tractable approaches. I show that under some conditions on the ϕ-mixing coefficients, a modified version of the UCB algorithm proves effective. The main challenge is that, unlike in the i.i.d. setting, the distributions of the sampled pay-offs may not have the same characteristics as those of the original bandit arms. In particular, the ϕ-mixing property does not necessarily carry over. This is overcome by carefully controlling the effect of a sampling policy on the pay-off distributions. Some of the proof techniques developed can be more generally used in the context of online sampling under dependence. Proposed algorithms are accompanied with corresponding regret analysis. I will ensure to make the talk accessible to non-experts. .

Fri, Nov 24, '17
15:00 - 16:00
CRiSM Seminar
A1.01
3-4pm A1.01, Nov 24, 2017 - Song Liu

Title: Trimmed Density Ratio Estimation

Abstract: Density ratio estimation has become a versatile tool in machine learning community recently. However, due to its unbounded nature, density ratio estimation is vulnerable to corrupted data points, which often pushes the estimated ratio toward infinity. In this paper, we present a robust estimator which automatically identifies and trims outliers. The proposed estimator has a convex formulation, and the global optimum can be obtained via subgradient descent. We analyze the parameter estimation error of this estimator under high-dimensional settings. Experiments are conducted to verify the effectiveness of the estimator.

Fri, Dec 8, '17
15:00 - 17:00
CRiSM Seminar
A1.01
3-4pm A1.01, Dec 8, 2017 - Richard Samworth

Title: High-dimensional changepoint estimation via sparse projection

Abstract: Changepoints are a very common feature of big data that arrive in the form of a data stream. We study high dimensional time series in which, at certain time points, the mean structure changes in a sparse subset of the co-ordinates. The challenge is to borrow strength across the co-ordinates to detect smaller changes than could be observed in any individual component series. We propose a two-stage procedure called inspect for estimation of the changepoints: first, we argue that a good projection direction can be obtained as the leading left singular vector of the matrix that solves a convex optimization problem derived from the cumulative sum transformation of the time series. We then apply an existing univariate changepoint estimation algorithm to the projected series. Our theory provides strong guarantees on both the number of estimated changepoints and the rates of convergence of their locations, and our numerical studies validate its highly competitive empirical performance for a wide range of data-generating mechanisms. Software implementing the methodology is available in the R package InspectChangepoint.

Fri, Jan 19, '18
14:00 - 15:00
CRiSM Seminar
MA_B1.01
2-3pm MA B1.01, 19 Jan, 2018 - Jonas Peters -

title + abstract TBA

Fri, Feb 2, '18
14:00 - 15:00
CRiSM Seminar
MA_B1.01
2-3pm MA B1.01, 2 Feb, 2018 - Robin Evans -

Title: Geometry and statistical model selection Abstract: TBA

 
15:00 - 16:00
CRiSM Seminar
A1.01
Fri, Feb 16, '18
14:00 - 15:00
CRiSM Seminar
MA_B1.01
2-3pm MA B1.01, 16 Feb, 2018 - TBC

title + abstract TBA

 
15:00 - 16:00
CRiSM Seminar
A1.01
Fri, Mar 2, '18
14:00 - 15:00
CRiSM Seminar
MA_B1.01
2-3pm MA B1.01, 2 March, 2018 - Judith Rousseau -

title + abstract TBA

Fri, Mar 16, '18
14:00 - 15:00
CRiSM Seminar
MA_B1.01
2-3pm MA B1.01, 16 March, 2018 - TBC

title + abstract TBA

Fri, May 4, '18
14:00 - 15:00
CRiSM Seminar
D1.07
2-3pm D1.07, May 4, 2018 - Wenyang Zhang -

title + abstract TBA

Fri, May 18, '18
14:00 - 15:00
CRiSM Seminar
D1.07
2-3pm D1.07, May 18, 2018 - TBC

title + abstract TBA

Fri, Jun 1, '18
14:00 - 15:00
CRiSM Seminar
D1.07
2-3pm D1.07, June 1, 2018 - Victor Panaretos -

title + abstract TBA

Fri, Jun 15, '18
14:00 - 15:00
CRiSM Seminar
D1.07
2-3pm D1.07, June 15, 2018 - TBC

title + abstract TBA

Fri, Jun 29, '18
14:00 - 15:00
CRiSM Seminar
D1.07
2-3pm D1.07, June 29, 2018 - TBC

title + abstract TBA