Coherent Knowledge Processing at Maximum Entropy by SPIRIT
نویسندگان
چکیده
SPIRIT is an expert system shell for probabilistic knowledge bases. Knowledge acquisition is performed by processing facts and rules on discrete variables in a rich syntax. The shell generates a probability distribution which respects all acquired facts and rules and which maximizes entropy. The user-friendly devices of SPIRIT to define variables, formulate rules and create the knowledge base are revealed in detail. Inductive learning is possible. Medium sized applications show the power of the system.
منابع مشابه
Probabilistic Knowledge Processing and Remaining Uncertainty
Information is indispensable in preparing economic decisions purposefully. In this paper knowledge is represented by a probability distribution. Knowledge acquisition is realized by the principle of maximum entropy which guarantees an unsophisticated expansion of knowledge and provides an unbiased basis for decision processes. Under a mathematical point of view entropy measures remaining uncert...
متن کاملCoherent Control of Quantum Entropy via Quantum Interference in a Four-Level Atomic System
The time evaluation of quantum entropy in a four-level double- type atomic system is theoretically investigated. Quantum entanglement of the atom and its spontaneous emission fields is then discussed via quantum entropy. It is found that the degree of entanglement can be increased by the quantum interference induced by spontaneous emission. The phase dependence of the atom-field entanglement is...
متن کاملEntropy-driven inference and inconsistency
Probability distributions on a set of discrete variables are a suitable means to represent knowledge about their respective mutual dependencies. When now things become evident such a distribution can be adapted to the new situation and hence submitted to a sound inference process. Knowledge acquisition and inference are here performed in the rich syntax of conditional events. Both, acquisition ...
متن کاملSequence Modeling with Mixtures of Conditional Maximum Entropy Distributions
We present a novel approach to modeling sequences using mixtures of conditional maximum entropy distributions. Our method generalizes the mixture of first-order Markov models by including the “long-term” dependencies in model components. The “long-term” dependencies are represented by the frequently used in the natural language processing (NLP) domain probabilistic triggers or rules (such as “A...
متن کاملMaximum - entropy theory of steady - state quantum trans - port
– We develop a new theoretical framework for describing steady-state quantum transport phenomena, based on the general maximum-entropy principle of non-equilibrium statistical mechanics. The general form of the many-body density matrix is derived, which contains the invariant part of the current operator that guarantees the non-equilibrium and steady-state character of the ensemble. Several exa...
متن کامل