tennsS v1.53 Reference Manual

Copyright (C) 1997 - 2002 Mike Arnold, Altjira Software, mikea@altjira.com

Go to the previous, next section, table of contents.


Tmonorule_Sig

Class Tmonorule_Sig derived from Netlearn

A single variable learning rule that can be bound to the underlying network. The rule is defined by the referenced <rule>. Its scope is defined in terms of a set of nodes together with a set of connections for each node. The rule is referenced and must be of type Particle. For cnset objects to be valid, each record must contain one node and any number of connections. The <rule> governing plasticity is applied once for each connection. The plastic <parameter> is defined relative to each connection. The single variable to the rule is given by calling <object> defined relative to the corresponding node. When calling each object, it is passed the network time. If <fix-at-zero> is true, then the parameters are bounded at zero, so that the sign of the parameters does not change.


Naming

rule references the rule

Configuration

rule=<ref> sets the rule fixed-delay=<bool> fixed delays flag (true) fix-at-zero=<bool> fix parameters at zero (true)

Additional arguments to existing commands

bind object=<id> [parameter=<id>] [rule=<id>] [fix-at-zero=<bool>] <further arguments for derived commands> binds itself to the underlying network structure object=<id> the single variable defined relative to the node parameter=<id> the plastic parameter relative to the connection rule=<id> the rule determining the plastic change in the parameter fix-at-zero=<bool> whether to preserve the sign of the parameters

Derived Classes


Go to the previous, next section, table of contents. Document generated Mon Jun 16 02:19:05 GMT 2003