alpha_v

Specifies the learning rate for v-values (the associative strength between stimulus element - behavior pairs), used in the updating of the v-values. See The mechanisms.

The parameter alpha_v can be set to certain values for certain particular element-behavior pairs. It can also be specified as a single value, in which case the same learning rate is used for all element-behavior pairs.

Syntax

alpha_v = e1->b1: v1, e1->b2: v2, ..., en->bn: vn, default: d
alpha_v = v1

where v1,v2,...,vn are scalar expressions.

Description

alpha_v = e1->b1: v1, e1->b2: v2, ..., en->bn: vn, default: d sets the learning rate for

  • v(e1->b1) to v1,

  • v(e1->b2) to v2, …,

  • v(en->bn) to v2,

and the learning rate for all other v-values to d.

  • The specification is independent of the list order:

    alpha_v = e1->b1:v1, e1->b2:v2, default:d

    is the same as

    alpha_v = e1->b2:v2, default:d, e1->b1:v1.

  • default need not be specified if all possible combinations element->behavior are present in the list. For example,

    stimulus_elements = e1, e2
behaviors = b1, b2
alpha_v = e1->b1:v11, e1->b2:v12, e2->b1:v21, e2->b2:v22

alpha_v = v1 sets the learning rate for all v-values to v1.

  • alpha_v = v1 is the same as alpha_v = default: v1.

Dependencies

  • The properties stimulus_elements and behaviors must be specified before alpha_v.

  • Each stimulus element used in the specification of alpha_v must be present in the parameter stimulus_elements.

  • Each behavior used in the specification of alpha_v must be present in the parameter behaviors.

  • Applicable in all mechanisms except the Rescorla-Wagner mechanism. (Use alpha_vss in Rescorla-Wagner.)

Examples

@variables x = 1
alpha_v = element1->behavior1: x, element1->behavior2: x+1, default:x+2

sets the learning rate for v(element1->behavior1) to 1 and for v(element1->behavior1) to 2, and for the remaining possible element-behavior pairs to 3.

alpha_v = element1->behavior1: 0.5, default:0

sets the learning rate for v(element1->behavior1) to 0.5, and for the remaining possible element-behavior pairs to 3.

alpha_v = 0.1

sets the learning rate for all v-values to 0.1.