Learning via Queries with Teams and Anomalies Authors: William I. Gasarch, Efim B. Kinber, Mark G. Pleszkoch, Carl H. Smith and Thomas Zeugmann Source: Fundamenta Informaticae 62, 1995, 67 - 89. Abstract. The paper studies the learning power of pluralistic inference alagorithms and of active learning devices that are allowed to ask questions. These learning models are studied in dependence on the degree of pluralism and in dependence on the query languages available. The query languages considered are all decidable first order languages. Three new hierarchies of more and more powerful query learners are established. The hierachies obtained are related to one another as well as to learning with anomalies. All these techniques increase the learning power but in very different directions. Finally, we consider the relaxation that the pluralistic and active learners are allowed to infer programs not exceeding an a priori fixed bound of errors. ©Copyright 1995, Fundamenta Informaticae