Next: Specifying Output Signal Weights, Previous: Repeated Sequence of Operations, Up: Adaptive Probabilistic Mapping [Contents][Index]

The efficiency of operation of a small or large actor, that is, the optimality of output signals the actor adaptively emits, depends on the form of a function returning the relative probability of an output signal. It may well be true that, for different kinds of applications, different functions are more suitable. QSMM provides a few built-in functions a developer may select for use by an actor. Unfortunately, the author of QSMM experiences problems with: (1) providing a sound theoretical basis for using those functions in QSMM (that basis might not exist for some functions); (2) devising functions that give better efficiency than the functions currently provided in QSMM. The author’s own experimenting had shown that, for some special cases, there do exist more efficient functions, but he did fail to find general forms for those functions making it possible to utilize them for wider ranges of values of actor parameters. To provide a way to solve these problems by developers, QSMM starting from version 1.15 supports specifying a custom relative probability function.

The main parameter of a relative probability function is its type. If the type specifies a custom form for the function, the developer shall provide a helper function for this form. Common relative probability function forms depend on the number of output signals and the way of spur perception and spur weight for every spur type. All relative probability function forms depend on actor temperature.

• Relative Probability Function Types | ||

• Helper Relative Probability Functions | ||

• Spur Perception | ||

• Spur Weight | ||

• Number of Output Signals | ||

• Actor Temperature |