Tuesday, October 18, 2005

minimal controller for circle-diamond discrimination task

here's an update on these experiments...

1. binary genotype added on top of the real-valued genotype so as to turn on/off internode connections and also stateless/statefull nodes. 30 binary genes in total. 5 for the 5 time-constants of the 5 inter-nodes which are modified to same value as the time-step of integration in the case of 0 and to the mapped value on the real in the case of 1. 25 values for the inter-node connections between the 5 inter-nodes. 0 corresponds to no connection and 1 corresponds to a connection with the strength of the real-valued mapped gene.
2. sensory and motor nodes have no state (time-constant = time-step of integration).

sets of experiments:
[a] gradual fitness: smaller weights and smaller time-constants improve fitness.
[b] threshold fitness: fitness only improves if inter-connection weights are off and time-constants are off - smaller doesn't count.

results so far:
out of 5 runs 1 was fully successful. it evolved the first 1000 generations with fitness function from set [a] and the last 100 with the fitness function from set [b] given that it needed to prune out only one more recurrent connection which was already very low (i.e. 0.48). (will add some figures soon!)

plan to leave 20 runs for fitness function [b] and cross my fingers.
that should be it.
hopefully at least a couple of reactive solutions evolve in which case i shall analyse them and report them to our friends at indiana ;)


At 9:49 pm, Blogger eduardo said...

update on the update:
100 experiments in total running.
5 sets. A: networks constrained to be purely reactive; B: networks constrained to be purely non-reactive; C: reactivity is evolvable with the binary genotype with selection pressure towards simpler (i.e. reactive) networks. C1: population starts off reactive; C2: population starts off purely non-reactive; C3: population starts off somewhere randomly in-between.

- have the steady state procedure ready to analyse how the behaviour is produced from the agent-environment interaction.

- want to implement the parameter to morph from circle to diamond to study the categorisation performance but are having some problems with it.

- reading the papers and writing the structure of a 'report'


Post a Comment

<< Home