In recent years the belief network has been used increasingly to model systems in AI that must pe... more In recent years the belief network has been used increasingly to model systems in AI that must perf orm uncertain inf erence. The de velopment of efficient algorithms fo r proba bilistic inf erence in belief networks has been a fo cus of much research in AI. Efficient al gorithms fo r certain classes of belief networks have been developed, but the problem of re porting the uncertainty in inf erred probabil ities has received little attention. A system should not only be capable of reporting the values of inf erred probabilities and/or the fa vorable choices of a decision; it should re port the range of possible error in the inf erred probabilities and/or c hoices. Two methods have been developed and im plemented fo r determining the variance in inf erred proba bilities in belief networks. These methods, the Approximate Propagation Method and the Monte Carlo Integration Method are dis cussed and compared in this paper.
In recent years the belief network has been used increasingly to model systems in AI that must pe... more In recent years the belief network has been used increasingly to model systems in AI that must perf orm uncertain inf erence. The de velopment of efficient algorithms fo r proba bilistic inf erence in belief networks has been a fo cus of much research in AI. Efficient al gorithms fo r certain classes of belief networks have been developed, but the problem of re porting the uncertainty in inf erred probabil ities has received little attention. A system should not only be capable of reporting the values of inf erred probabilities and/or the fa vorable choices of a decision; it should re port the range of possible error in the inf erred probabilities and/or c hoices. Two methods have been developed and im plemented fo r determining the variance in inf erred proba bilities in belief networks. These methods, the Approximate Propagation Method and the Monte Carlo Integration Method are dis cussed and compared in this paper.
In recent years the belief network has been used increasingly to model systems in AI that must pe... more In recent years the belief network has been used increasingly to model systems in AI that must perf orm uncertain inf erence. The de velopment of efficient algorithms fo r proba bilistic inf erence in belief networks has been a fo cus of much research in AI. Efficient al gorithms fo r certain classes of belief networks have been developed, but the problem of re porting the uncertainty in inf erred probabil ities has received little attention. A system should not only be capable of reporting the values of inf erred probabilities and/or the fa vorable choices of a decision; it should re port the range of possible error in the inf erred probabilities and/or c hoices. Two methods have been developed and im plemented fo r determining the variance in inf erred proba bilities in belief networks. These methods, the Approximate Propagation Method and the Monte Carlo Integration Method are dis cussed and compared in this paper.
In recent years the belief network has been used increasingly to model systems in AI that must pe... more In recent years the belief network has been used increasingly to model systems in AI that must perf orm uncertain inf erence. The de velopment of efficient algorithms fo r proba bilistic inf erence in belief networks has been a fo cus of much research in AI. Efficient al gorithms fo r certain classes of belief networks have been developed, but the problem of re porting the uncertainty in inf erred probabil ities has received little attention. A system should not only be capable of reporting the values of inf erred probabilities and/or the fa vorable choices of a decision; it should re port the range of possible error in the inf erred probabilities and/or c hoices. Two methods have been developed and im plemented fo r determining the variance in inf erred proba bilities in belief networks. These methods, the Approximate Propagation Method and the Monte Carlo Integration Method are dis cussed and compared in this paper.
Uploads
Papers by Peter Che