A wide variety of algorithms have been studied that fall into these categories.<ref name="settles" /><ref name="olsson" />
While the traditional AL strategies can achieve remarkable performance, it is often challenging to predict which strategy is the most suitable in a particular situation. In recent years, meta-learning algorithms have been gaining in popularity. Some of them have been proposed to tackle the problem of learning AL strategies instead of relying on manually designed strategies.<ref>{{cite book|last1=Konyushkova|first1=K. |last2=Sznitman|first2=R.|last3=Fua|first3=P.|chapter=Learning Active Learning from Data |date=2017|title=Advances in Neural Information Processing Systems|arxiv=1703.03365}}</ref><ref>{{cite book|last1=Desreumaux |first1=Louis |last2=Lemaire|first2=Vincent|chapter=Learning Active Learning at the Crossroads? Evaluation and Discussion|date=2020|title=Interactive Adaptive Learning, ECML-PKDD, 2020|arxiv=2012.09631}}</ref><ref>{{cite journal |last1=Konyushkova|first1=K. |last2=Sznitman|first2=R.|last3=Fua|first3=P.|title=Discovering General-Purpose Active Learning Strategies |date=2019|arxiv=1810.04114}}</ref>
