Titlebook: Machine Learning: ECML‘97; 9th European Confere Maarten Someren,Gerhard Widmer Conference proceedings 1997 Springer-Verlag Berlin Heidelber

Foregery

黑豹

Metrics on terms and clauses,t is easy to form metrics on clauses, by a technique due to F.Hausdorff. They will be applicable to generalization from sets of ground clauses, to discovery of heuristic guidance for theorem proving, and to inductive logic programming.

monologue

魯莽

Compression-based pruning of decision lists,ains. Simple hill-climbing search for the most compressive decision list significantly reduces the complexity of a given decision list while not impeding and sometimes even improving its predictive accuracy.

Terrace

平躺

Induction of feature terms with INDIE,. An example of such inductive methods, ., is presented. . performs bottom-up heuristic search on the subsumption lattice of the feature term space. Results of this method on several domains are explained.

暫停,間歇

,θ-subsumption for structural matching,such that the remaining problems disappear. Defining clauses as universally quantified disjunctions of literals and productions as existentially quantified conjunctions of literals, it is shown that the lattice on clauses imposed by θ-subsumption is order-isomorphic to the lattice on productions needed for structural matching.

innovation

Parallel and distributed search for structure in multivariate time series,ultivariate time series, and that the distributed version finds the same structure in approximately 1/. of the time required by ., where . is the number of machines across which the search is distributed.

Kernel

Integrated learning and planning based on truncating temporal differences,entially no additional computational expense. Two different ways of combining TTD with planning are proposed which make it possible to benefit from λ>0 in both the learning and planning processes. The algorithms are evaluated experimentally on a family of grid path-finding tasks and shown to indeed

堅(jiān)毅