Previous work assigned debaters/consultants an answer to argue for.

Despite the rich potential for innovation, the learning process of algorithms under real-world conditions is a challenge [1, 38, 46].

Infants possess a remarkable ability to rapidly learn and process visual inputs.

In this work we study the problem of actively learning binary classifiers from a given concept class, i.e., learning by utilizing unlabeled data and submitting targeted queries about their labels to a domain expert. We evaluate the quality of our solutions by considering the learning curves they induce, i.e., the rate of

A.1 Deriving the Optimum of the KL-Constrained Reward Maximization Objective In this appendix, we will derive Eq. 4. Analogously to Eq. 3, we optimize the following objective: max

We conduct extensive experiments on various benchmark datasets, empirically demonstrate the effectiveness of Terra.