Sebastian's slow blog

  • How to report uncertainty

    Sebastian Nowozin - Thu 19 March 2015 -

    Error bars and the \(\pm\)-notation are used to quantitatively convey uncertainty in experimental results. For example, you would often read statements like \(140.7 \textrm{Hz} \pm 2.8 \textrm{ Hz SEM}\) in a paper to report both an …


  • Estimating Discrete Entropy, Part 3

    Sebastian Nowozin - Sat 07 March 2015 -

    In the last two parts (part one, part two) we looked at the problem of entropy estimation and several popular estimators.

    In this final article we will take a look at two Bayesian approaches to the problem.

    Bayesian Estimator due …


  • Machine Learning in Cambridge 2015

    Sebastian Nowozin - Thu 26 February 2015 -

    This year we (Zoubin, together with David and myself) are again organizing a workshop event for the local Cambridge (UK) machine learning community. The schedule is available at the workshop homepage, Machine Learning in Cambridge 2015, and we also plan …


  • Estimating Discrete Entropy, Part 2

    Sebastian Nowozin - Sat 21 February 2015 -

    In the last part we have looked at the basic problem of discrete entropy estimation. In this article we will see a number of proposals of improved estimators.

    Miller Correction

    In 1955 George Miller proposed a simple correction to the …


  • Estimating Discrete Entropy, Part 1

    Sebastian Nowozin - Sat 07 February 2015 -

    Estimation of the entropy of a random variable is an important problem that has many applications. If you can estimate entropy accurately, you can also estimate mutual information, which allows you to find dependent random variables in large data sets …