Computational Aspects of Parallel Attribute-Efficient Learning

We address the problem of nonadaptive learning of Boolean functions with few relevant variables by membership queries. In another recent paper [7] we have characterized those assignment families (query sets) which are sufficient for nonadaptive learning of this function class, and we studied the query number. However, the reconstruction of the given Boolean function from the obtained responses is an important matter as well in applying such nonadaptive strategies. The computational amount for this is apparently too high if we use our query families in a straightforward way. Therefore we introduce algorithms where also the computational complexity is reasonable, rather than the query number only. The idea is to apply our assignment families to certain coarsenings of the given Boolean function, followed by simple search and verification routines.

[1]  Emanuel Knill,et al.  A Comparative Survey of Non-Adaptive Pooling Designs , 1996 .

[2]  Tibor Hegedüs.  Combinatorial Results on the Complexity of Teaching and Learning , 1994, MFCS.

[3]  Gadiel Seroussi,et al.  Vector sets for exhaustive testing of logic circuits , 1988, IEEE Trans. Inf. Theory.

[4]  Rajeev Motwani,et al.  Randomized Algorithms , 1995, SIGA.

[5]  Lisa Hellerstein,et al.  Learning in the presence of finitely or infinitely many irrelevant attributes , 1991, COLT '91.

[6]  Peter Damaschke.  Adaptive versus nonadaptive attribute-efficient learning , 1998, STOC '98.

[7]  Nader H. Bshouty,et al.  On the exact learning of formulas in parallel , 1992, Proceedings., 33rd Annual Symposium on Foundations of Computer Science.

[8]  David C. Torney,et al.  Optimal Pooling Designs with Error Detection , 1994, J. Comb. Theory, Ser. A.

[9]  Lisa Hellerstein,et al.  PAC learning with irrelevant attributes , 1994, Proceedings 35th Annual Symposium on Foundations of Computer Science.

[10]  Piotr Indyk,et al.  On Learning Disjunctions of Zero-One Treshold Functions with Queries , 1997, ALT.

[11]  N. Littlestone.  Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[12]  Robert E. Schapire,et al.  Exact Identification of Read-Once Formulas Using Fixed Points of Amplification Functions , 1993, SIAM J. Comput..

[13]  C. Colbourn,et al.  The CRC handbook of combinatorial designs , edited by Charles J. Colbourn and Jeffrey H. Dinitz. Pp. 784. $89.95. 1996. ISBN 0-8493-8948-8 (CRC). , 1997, The Mathematical Gazette.

[14]  Sampath Kannan,et al.  Group testing problems with sequences in experimental molecular biology , 1997, Proceedings. Compression and Complexity of SEQUENCES 1997 (Cat. No.97TB100171).

[15]  Aravind Srinivasan,et al.  Splitters and near-optimal derandomization , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[16]  Lisa Hellerstein,et al.  Attribute-efficient learning in query and mistake-bound models , 1996, COLT '96.

[17]  Michael Kearns,et al.  On the complexity of teaching , 1991, COLT '91.