Connectionist learning models have had considerable empirical success, but it is hard to characterize exactly what they learn. The learning of finite-state languages (FSL) from example strings is a domain which has been extensively studied and might provide an opportunity to help understand connectionist learning. A major problem is that traditional FSL learning assumes the storage of all examples and thus violates connectionist principles. This paper presents a provably correct algorithm for inferring any minimum-state deterministic finite-state automata (FSA) from a complete ordered sample using limited total storage and without storing example strings. The algorithm is an iterative strategy that uses at each stage a current encoding of the data considered so far, and one single sample string. One of the crucial advantages of our algorithm is that the total amount of space, used in the course oflearning, for encoding any finite prefix of the sample is polynomial in the size of the inferred minimum state deterministic FSA. The algorithm is also relatively efficient in time and has been implemented. More importantly, there is a connectionist version of the algorithm that preserves these properties. The connectionist version requires much more structure than the usual models and has not yet been implemented. But it does significantly extend the scope of connectionist learning systems and helps relate them to other paradigms. We also show that no machine with finite working storage can identify iteratively the FSL from arbitrary presentations.The Xerox Corporation University Grants Program provided equipment used in the preparation of this paper.

Alztmct: The notion of ~fa~" derlvat~on in a term-rewriting system is introduced, whereby every rewrite rule evu~led i~#f~z~tsll/oft9Tt along a derivaiion is infinitelyoften applied. A term-rewriting s y s t e m is f c r / r / I / -t e r m i r t a /~ iff all its fair derivations are finite.

No abstract

This first-ever study of the use of restraint on psychiatric patients in Israel sought to determine the dominant motivations in the decision to use restraint, which patients were most likely to be restrained, and whether there was a consistent policy on the matter. A survey of the official records of every instance of restraint during one month in the closed wards of all government psychiatric hospitals, supplemented with interviews, revealed that 14.2% of the study population had undergone restraint. The declared reasons were conventional, i.e. violence, disturbed behaviour, etc. Undeclared was an interaction between patients and staff, both the most professional and the less skilled, which surprised the authors and requires more investigation. Some subgroups, e.g. women and certain immigrant groups, were restrained markedly more frequently than other groups. No consistency of policy was found. Overall, much of the restraint applied is deemed unnecessary and recommendations are made for its reduction.

Connectionist learning models have had considerable empirical success, but it is hard to characterize exactly what they learn. The learning of finite-state languages (FSL) from example strings is a domain which has been extensively studied and might provide an opportunity to help understand connectionist learning. A major problem is that traditional FSL learning assumes the storage of all examples and thus violates connectionist principles. This paper presents a provably correct algorithm for inferring any minimum-state deterministic finite-state automata (FSA) from a complete ordered sample using limited total storage and without storing example strings. The algorithm is an iterative strategy that uses at each stage a current encoding of the data considered so far, and one single sample string. One of the crucial advantages of our algorithm is that the total amount of space, used in the course oflearning, for encoding any finite prefix of the sample is polynomial in the size of the inferred minimum state deterministic FSA. The algorithm is also relatively efficient in time and has been implemented. More importantly, there is a connectionist version of the algorithm that preserves these properties. The connectionist version requires much more structure than the usual models and has not yet been implemented. But it does significantly extend the scope of connectionist learning systems and helps relate them to other paradigms. We also show that no machine with finite working storage can identify iteratively the FSL from arbitrary presentations.The Xerox Corporation University Grants Program provided equipment used in the preparation of this paper.