초록

Neural network architectures are represented by symbol strings. An initial population of networks is trained and evaluated. The strings representing the fittest networks are modified according to a genetic algorithm and the process is repeated until an optimized network is produced.

대표청구항 ▼

Neural network architectures are represented by symbol strings. An initial population of networks is trained and evaluated. The strings representing the fittest networks are modified according to a genetic algorithm and the process is repeated until an optimized network is produced. llocate the port

Neural network architectures are represented by symbol strings. An initial population of networks is trained and evaluated. The strings representing the fittest networks are modified according to a genetic algorithm and the process is repeated until an optimized network is produced. llocate the portion of the total-usage measure after a given time duration. 5. The provider as claimed in claim 1, wherein: the security device further encodes the at least one baseline-usage parameter in a secure form, the secure form including at least one of: a digital signature bound to the at least one baseline-usage parameter, and an encryption of the at least one baseline-usage parameter. 6. A method of providing content material, the method comprising the steps: recording the content material on a recording medium; allocating a portion of a total-usage measure to the content material on the recording medium; determining a measure of usage associated with the recording medium; and recording at least one baseline-usage parameter on the recording medium based on the measure of usage associate with the recording medium and the portion of the total-usage measure, to facilitate a determination of a validity period associated with the content material on the recording medium, wherein said method further comprises the step: binding the at least one baseline-usage parameter to a ticket that is a hashed value of a watermark that is associated with the content material. 7. The method as claimed in claim 6, wherein the method further comprises the step: de-allocating the portion of the total-usage measure when the recording medium is subsequently processed. 8. The method as claimed in claim 6, wherein the method further comprises the step: de-allocating the portion of the total-usage measure after a given time duration. 9. The method as claimed in claim 6, wherein the method further comprises the step: binding the at least one baseline-usage parameter to the content material. 10. The method as claimed in claim 6, wherein the method further comprises the step: encoding the at least one baseline-usage parameter in a secure form, the secure form including at least one of: a digital signature bound to the at least one baseline-usage parameter, and an encryption of the at least one baseline-usage parameter.