AI systems must be able to learn, reason logically, and handle uncertainty. While much research has focused on each of these goals individually, only recently have we begun to attempt to achieve all three at once. In ...
详细信息
It is well known by inductivelogicprogramming (ilp) practioners that ilp systems usually take a long time to find valuable models (theories). the problem is specially critical for large datasets, preventing ilp syst...
详细信息
We propose a new approach to inductivelogicprogramming i that systematically exploits caching and offers a number of advantages over current systems. It avoids redundant computation, is more amenable to the use of s...
详细信息
In this paper, we focus on the problem of learning reactive skills for use by physical agents. We propose a new representation for such procedures, teleoreactive logic programs, along with an interpreter that utilizes...
详细信息
ISBN:
(纸本)3540281770
In this paper, we focus on the problem of learning reactive skills for use by physical agents. We propose a new representation for such procedures, teleoreactive logic programs, along with an interpreter that utilizes them to achieve goals. After this, we describe a learning method that acquires these structures in a cumulative manner through problem solving. We report experiments in three domains that involve multiple levels of skilled behavior. We also review related work and discuss directions for future research.
this paper studies equivalence issues in inductivelogicprogramming. A background theory B-1 is inductively equivalent to another background theory B-2 if B-1 and B-2 induce the same hypotheses for any given set of e...
详细信息
ISBN:
(纸本)3540281770
this paper studies equivalence issues in inductivelogicprogramming. A background theory B-1 is inductively equivalent to another background theory B-2 if B-1 and B-2 induce the same hypotheses for any given set of examples. inductive equivalence is useful to compare inductive capabilities among agents having different background theories. Moreover, it provides conditions for optimizing background theories through appropriate program transformations. In this paper, we consider three different classes of background theories: clausal theories, Horn logic programs, and nonmonotonic extended logic programs. We show that logical equivalence is the necessary and sufficient condition for inductive equivalence in clausal theories and Horn logic programs. In nonmonotonic extended logic programs, on the other hand, strong equivalence is necessary and sufficient for inductive equivalence in general. Interestingly, however, we observe that several existing induction algorithms require weaker conditions of equivalence under restricted problem settings. We also discuss connection to equivalence in abductive logic and conclude that the notion of strong equivalence is useful to characterize equivalence of non-deductive reasoning.
logical Bayesian Networks (LBNs) have recently been introduced as another language for knowledge based model construction of Bayesian networks, besides existing languages such as Probabilistic Relational Models (PRMs)...
详细信息
ISBN:
(纸本)3540281770
logical Bayesian Networks (LBNs) have recently been introduced as another language for knowledge based model construction of Bayesian networks, besides existing languages such as Probabilistic Relational Models (PRMs) and Bayesian logic Programs (BLPs). the original description of LBNs introduces them as a variant of BLPs and discusses the differences with BLPs but still leaves room for a deeper discussion of the relationship between LBNs and BLPs. Also the relationship to PRMs was not treated in much detail. In this paper, we first give a more compact and clear definition of LBNs. Next, we describe in more detail how PRMs and BLPs relate to LBNs. Like this we not only see what the advantages and disadvantages of LBNs are with respect to PRMs and BLPs, we also gain more insight into the relationships between PRMs and BLPs.
the handling of exceptions in multiclass problems is a tricky issue in inductivelogicprogramming (ilp). In this paper we propose a new formalization of the ilp problem which accounts for default reasoning, and is en...
详细信息
the handling of exceptions in multiclass problems is a tricky issue in inductivelogicprogramming (ilp). In this paper we propose a new formalization of the ilp problem which accounts for default reasoning, and is encoded with first-order possibilistic logic. We show that this formalization allows us to handle rules with exceptions, and to prevent an example to be classified in more than one class. the possibilistic logic view of ilp problem, can be easily handled at the algorithmic level as an optimization problem.
Frequent closed pattern discovery is one of the most important topics in the studies of the compact representation for data mining. In this paper, we consider the frequent closed pattern discovery problem for a class ...
详细信息
ISBN:
(纸本)3540281770
Frequent closed pattern discovery is one of the most important topics in the studies of the compact representation for data mining. In this paper, we consider the frequent closed pattern discovery problem for a class of structured data, called attribute trees (AT), which is a subclass of labeled ordered trees and can be also regarded as a fragment of description logic with functional roles only. We present an efficient algorithm for discovering all frequent closed patterns appearing in a given collection of attribute trees. By using a new enumeration method, called the prefix-preserving closure extension, which enable efficient depth-first search over all closed patterns without duplicates, we show that this algorithm works in polynomial time both in the total size of the input database and the number of output trees generated by the algorithm. To our knowledge, this is one of the first result for output-sensitive algorithms for frequent closed substructure disocvery from trees and graphs.
this paper illustrates the use of inductivelogicprogramming to program agents that learn rules of behaviour from simulated histories of their embedding systems. We have shown how a ilp system can be used to learn ru...
详细信息
ISBN:
(纸本)0889865248
this paper illustrates the use of inductivelogicprogramming to program agents that learn rules of behaviour from simulated histories of their embedding systems. We have shown how a ilp system can be used to learn rules in a representation very close to the one used to guide the simulation of a multi-agent system. this establishes the feasibility of embedding (resource- bounded) learners as agents that take part in simulating a complex system.
暂无评论