=Paper=
{{Paper
|id=Vol-2660/ialatecml_invitedtalk3
|storemode=property
|title=How to Measure Uncertainty in Uncertainty Sampling for Active Learning
|pdfUrl=https://ceur-ws.org/Vol-2660/ialatecml_invitedtalk3.pdf
|volume=Vol-2660
|authors=Eyke Hüllermeier
|dblpUrl=https://dblp.org/rec/conf/pkdd/Hullermeier20
}}
==How to Measure Uncertainty in Uncertainty Sampling for Active Learning==
<pdf width="1500px">https://ceur-ws.org/Vol-2660/ialatecml_invitedtalk3.pdf</pdf>
<pre>
 How to measure uncertainty in Uncertainty
       Sampling for Active Learning

                                Eyke Hüllermeier

                         University Potsdam, Germany
                           eyke@uni-paderborn.de



    Abstract. Various strategies for active learning have been proposed in
    the machine learning literature. In uncertainty sampling, which is among
    the most popular approaches, the active learner sequentially queries the
    label of those instances for which its current prediction is maximally
    uncertain. The predictions as well as the measures used to quantify the
    degree of uncertainty, such as entropy, are traditionally of a probabilistic
    nature. Yet, alternative approaches to capturing uncertainty in machine
    learning, alongside with corresponding uncertainty measures, have been
    proposed in recent years. In particular, some of these measures seek to
    distinguish different sources and to separate different types of uncer-
    tainty, such as the reducible (epistemic) and the irreducible (aleatoric)
    part of the total uncertainty in a prediction. This talk elaborates on
    the usefulness of such measures for uncertainty sampling and compares
    their performance in active learning. To this end, uncertainty sampling
    is instantiated with different measures, the properties of the sampling
    strategies thus obtained are analyzed and compared in an experimental
    study.




© 2020 for this paper by its authors. Use permitted under CC BY 4.0.

</pre>