22 episodes

The Department of Statistics at Oxford is a world leader in research including computational statistics and statistical methodology, applied probability, bioinformatics and mathematical genetics. In the 2014 Research Excellence Framework (REF), Oxford's Mathematical Sciences submission was ranked overall best in the UK.
This is an exciting time for the Department. We have now moved into our new home on St Giles and we are currently settling in.

The new building provides improved lecture and teaching space, a variety of interaction areas, and brings together researchers in Probability and Statistics. It has created a highly visible centre for the Department in Oxford.

Since 2010, the Department has been awarded over forty research grants with a total value of £9M, not counting several very large EPSRC and MRC funded awards for Centres for doctoral training.The main sponsors are the European Commission, EPSRC, the Medical Research Council and the Wellcome Trust.

We offer an undergraduate degree (BA or MMath) in Mathematics and Statistics, jointly with the Mathematical Institute.

At postgraduate level there is an MSc course in Applied Statistics, as well as a lively and stimulating environment for postgraduate research (DPhil or MSc by Research). Our graduates are employed in a wide range of occupational sectors throughout the world, including the university sector.

The Department co-hosts the EPSRC and MRC Centre for Doctoral Training (CDT) in Next-Generational Statistical Science- the Oxford-Warwick Statistics Programme OxWaSP.

Department of Statistics Oxford University

    • Education

The Department of Statistics at Oxford is a world leader in research including computational statistics and statistical methodology, applied probability, bioinformatics and mathematical genetics. In the 2014 Research Excellence Framework (REF), Oxford's Mathematical Sciences submission was ranked overall best in the UK.
This is an exciting time for the Department. We have now moved into our new home on St Giles and we are currently settling in.

The new building provides improved lecture and teaching space, a variety of interaction areas, and brings together researchers in Probability and Statistics. It has created a highly visible centre for the Department in Oxford.

Since 2010, the Department has been awarded over forty research grants with a total value of £9M, not counting several very large EPSRC and MRC funded awards for Centres for doctoral training.The main sponsors are the European Commission, EPSRC, the Medical Research Council and the Wellcome Trust.

We offer an undergraduate degree (BA or MMath) in Mathematics and Statistics, jointly with the Mathematical Institute.

At postgraduate level there is an MSc course in Applied Statistics, as well as a lively and stimulating environment for postgraduate research (DPhil or MSc by Research). Our graduates are employed in a wide range of occupational sectors throughout the world, including the university sector.

The Department co-hosts the EPSRC and MRC Centre for Doctoral Training (CDT) in Next-Generational Statistical Science- the Oxford-Warwick Statistics Programme OxWaSP.

    • video
    A primer on PAC-Bayesian learning *followed by* News from the PAC-Bayes frontline

    A primer on PAC-Bayesian learning *followed by* News from the PAC-Bayes frontline

    Benjamin Guedj, University College London, gives a OxCSML Seminar on 26th March 2021. Abstract: PAC-Bayes is a generic and flexible framework to address generalisation abilities of machine learning algorithms. It leverages the power of Bayesian inference and allows to derive new learning strategies. I will briefly present the key concepts of PAC-Bayes and highlight a few recent contributions from my group.

    • 59 min
    • video
    Approximate Bayesian computation with surrogate posteriors

    Approximate Bayesian computation with surrogate posteriors

    Julyan Arbel (Inria Grenoble - Rhône-Alpes), gives an OxCSML Seminar on Friday 30th April 2021, for the Department of Statistics.

    • 56 min
    • video
    Introduction to Bayesian inference for Differential Equation Models Using PINTS

    Introduction to Bayesian inference for Differential Equation Models Using PINTS

    Ben Lambert, Department of Computer Science, University of Oxford, gives the Graduate Lecture on Thursday 6th May 2021, for the Department of Statistics.

    • 57 min
    • video
    On classification with small Bayes error and the max-margin classifier

    On classification with small Bayes error and the max-margin classifier

    Professor Sara Van de Geer, ETH Zürich, gives the Distinguished Speaker Seminar on Thursday 29th April 2021 for the Department of Statistics.

    • 1 hr
    • video
    Convergence of Online SGD under Infinite Noise Variance, and Non-convexity

    Convergence of Online SGD under Infinite Noise Variance, and Non-convexity

    Murat Erdogdu gives the OxCSML Seminar on Friday 12th March, 2021, for the Department of Statistics.

    • 1 hr
    • video
    Distribution-dependent generalization bounds for noisy, iterative learning algorithms

    Distribution-dependent generalization bounds for noisy, iterative learning algorithms

    Karolina Dziugaite (Element AI), gives the OxCSML Seminar on 26th February 2021. Abstract: Deep learning approaches dominate in many application areas. Our understanding of generalization (relating empirical performance to future expected performance) is however lacking. In some applications, standard algorithms like stochastic gradient descent (SGD) reliably return solutions with low test error. In other applications, these same algorithms rapidly overfit. There is, as yet, no satisfying theory explaining what conditions are required for these common algorithms to work in practice. In this talk, I will discuss standard approaches to explaining generalization in deep learning using tools from statistical learning theory, and present some of the barriers these approaches face to explaining deep learning. I will then discuss my recent work (NeurIPS 2019, 2020) on information-theoretic approaches to understanding generalization of noisy, iterative learning algorithms, such as Stochastic Gradient Langevin Dynamics, a noisy version of SGD.

    • 54 min

Top Podcasts In Education

Listeners Also Subscribed To

More by Oxford University