This paper deals with the UAV learning model of Goldman, Kwek aud Scott [7], where "UAV" is the acronym for "Unspecified Attribute Values". As in [7], we consider exact learuing withiu the UAV framework. A smooth transition between exact learuing in the UAV settiug and standard exact learning is obtained by putting a fixed bound T 011 the number of uuspecified attribute values per instance. For r = 0, we obtain the standard model. For T = n (the total number of attributes), we obtaiu the (unrestricted) UAV model. Between these extremes, we fiud the hierarchies (UAV-MQ,)op
This paper deals with the UAV learning model of Goldman, Kwek, and Scott, where``UAV'' is the acronym for``unspecified attribute values.'' As they, we consider exact earning within the UAV framework. A smooth transition between exact learning in the UAV setting and standard exact learning is obtained by putting a fixed bound r on the number of unspecified attribute values per instance. For r=0, we obtain the standard model. For r=n (the total number of attributes), we obtain the (unrestricted) UAV model. Between these extremes, we find the hierarchies (UAV-MQ r ) 0 r n , (UAV-EQ r ) 0 r n , and (UAV-ARB-EQ r ) 0 r n . Our main results are as follows. We present various lower bounds on the number of ARB-EQs and UAV-MQs in terms of the Vapnik Chervonenkis dimension of the concept class. We show, furthermore, that a natural extension of Angluin's Sunflower Lemma is still applicable in the exact UAV learning model. Our UAV Sunflower Lemma allows the establishment of exponentially large lower bounds on the necessary number of UAV-MQs for several popular concept classes. On the other hand, we can show that slight simplifications of these classes are efficiently learnable using only a few UAV-MQs. Finally, we investigate the inherent structure of the aforementioned three hierarchies and the relations between them. It turns out that query type UAV-EQ r&1 is strictly stronger than UAV-EQ r (for each constant r). The analogous result for UAV-ARB-EQ is valid. Furthermore, UAV-MQ r+|(log n) is strictly stronger than UAV-MQ r . We also determine the relation between query types chosen from different hierarchies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.