초록 ▼

A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each o

A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each of a plurality of circuits is coupled to one of the plurality of neuron transfer function circuits and constructed to generate a derivative of the output. A weight update circuit updates the charge storage circuits based upon output from the plurality of transfer function circuits and output from the plurality of derivative circuits. In preferred embodiments, separate training and validation networks share the same set of charge storage circuits and may operate concurrently. The validation network has a separate transfer function circuits each being coupled to the charge storage circuits so as to replicate the training network's coupling of the plurality of charge storage to the plurality of transfer function circuits. The plurality of transfer function circuits may be constructed each having a transconductance amplifier providing differential currents combined to provide an output in accordance with a transfer function. The derivative circuits may have a circuit constructed to generate a biased differential currents combined so as to provide the derivative of the transfer function.

대표청구항 ▼

A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each o

A neural network circuit is provided having a plurality of circuits capable of charge storage. Also provided is a plurality of circuits each coupled to at least one of the plurality of charge storage circuits and constructed to generate an output in accordance with a neuron transfer function. Each of a plurality of circuits is coupled to one of the plurality of neuron transfer function circuits and constructed to generate a derivative of the output. A weight update circuit updates the charge storage circuits based upon output from the plurality of transfer function circuits and output from the plurality of derivative circuits. In preferred embodiments, separate training and validation networks share the same set of charge storage circuits and may operate concurrently. The validation network has a separate transfer function circuits each being coupled to the charge storage circuits so as to replicate the training network's coupling of the plurality of charge storage to the plurality of transfer function circuits. The plurality of transfer function circuits may be constructed each having a transconductance amplifier providing differential currents combined to provide an output in accordance with a transfer function. The derivative circuits may have a circuit constructed to generate a biased differential currents combined so as to provide the derivative of the transfer function. itor; (B) a mechanism constructed and adapted to determine a second score based on an amount owed by the debtor only to the selected creditor, wherein the second score is further based on at least one of the following tailored characteristics from a generic database: average months in file; number of trade lines last reported over 90 days delinquent; percentage amount now past due of maximum historic past due; total trade line balance; net fraction of subscriber's accounts; and total balance on subscriber's trade lines; and (C) a mechanism constructed and adapted to determine a collection score based on the first and second scores; wherein the collection score is further based on tailored characteristics for a generic database, and: (D) a mechanism constructed and adapted to determine the likely repayment amount based on the collection score and the amount owed to the selected creditor. 14. A computer-based system for predicting a likely repayment amount from a debtor who owes an amount to a selected creditor, the system comprising: (A) a mechanism constructed and adapted to determine a first score based on amounts owed by the debtor only to creditors other than the selected creditor; (B) a mechanism constructed and adapted to determine a second score based on an amount owed by the debtor only to the selected creditor; (C) a mechanism constructed and adapted to determine a collection score based on the first and second scores, wherein the collection score is further based on tailored characteristics from a generic database comprising: average months in file; number of trade lines last reported over 90 days delinquent; percentage amount now past due of maximum historic past due; total trade line balance; net fraction of subscriber's accounts; and total balance on subscriber'trade lines; and (D) a mechanism constructed and adapted to determine the likely repayment amount based on the collection score and the amount owed to the selected creditor. 15. A computer-readable medium with program instructions for predicting a likely repayment amount from a debtor who owes an amount to a selected creditor, the program comprising a method, which, when executed on a computer system, causes the compute