W. W. Cohen, The dual DFA learning problem: Hardness results for programming by demonstration and learning first-order representations, in: Proc. 9th Ann. Workshop on Computational Learning Theory (ACM, 1996) 29--40. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Prediction-Hardness of Acyclic Conjunctive Queries - Hirata (2000) (Correct)
....can receive a fact as a ground clause, Kietz [22, 23] implicitly has shown that acyclic conjunctive queries consisting of literals with at most 8 j ary predicate symbols (j # 2) are not pac learnable unless RP = PSPACE, without databases as background knowledge. Under the same setting, Cohen [8] has strengthened this result not to be predictable under the cryptographic assumptions. First, we obtain the following theorem. Note that the following proof is motivated by Cohen (Theorem 5 in [6] and Theorem 9 in [7] Theorem 5 For each j # 3, ACQ[j B] is not predictable from a simple ....
W. W. Cohen, The dual DFA learning problem: Hardness results for programming by demonstration and learning first-order representations, in: Proc. 9th Ann. Workshop on Computational Learning Theory (ACM, 1996) 29--40.
On the Hardness of Learning Acyclic Conjunctive Queries - Hirata (2000) (Correct)
....can receive an example as a ground clause, Kietz [20, 21] implicitly has shown that acyclic conjunctive queries consisting of literals with at most j ary predicate symbols (j # 2) are not pac learnable unless RP = PSPACE, without databases as background knowledge. Under the same setting, Cohen [8] has strengthened this result that they are not polynomially predictable under the cryptographic assumptions. On the other hand, by using Cohen s result (Theorem 3 in [6] we can claim that, for each j # 3, the recursive version of ACQ[j B] is not polynomially predictable from an extended ....
Cohen, W. W.: The dual DFA learning problem: Hardness results for programming by demonstration and learning first-order representations, Proc. 9th COLT, 29--40, 1996.
Learning from Ambiguity - Maron (1998) (17 citations) (Correct)
....true positive in each bag. As would be expected, local maxima present a major di#culty for the algorithm. A classic Machine Learning problem is learning a Deterministic Finite State Automata (DFA) from a series of strings which are labeled according to whether they are accepted by the DFA or not. Cohen, 1996 ] examines the dual of this problem: given a sequence of DFAs, each labeled positive or negative, find a string that is accepted by positive examples and not by negative ones. This is an ambiguous learning problem because a labeled example (DFA) represents a variety of di#erent strings. ....
William W. Cohen. The dual dfa learning problem: Hardness results for programming by demonstration and learning first-order representations. In Proceedings of the 1996 Conference on Computational Learning Theory, 1996.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC