Thursday, September 29, 2005

bits and pieces on ecal2005

I have to mention that there was quite a good reception in ECAL for (a) the paper in Tuci's workshop on evolving learning in the imprinting scenario, (b) the activate.d workshop, in particular the roundtable discussion and (c) the paper in collaboration with Ezequiel in the main conference on whether an agent which is embodied can ever be purely reactive. A couple of things about this last one....

I have been thinking more and more about the implications of the examples shown and discussed in that paper as well as a number of other similar examples. I will mention here two examples that I have in mind.

The first one is very straightforward and involves the use of inertia: catching a circle despite going blind after it starts falling. Beer suggested in his 1996 paper (Toward the evolution of dynamical neural networks for minimally cognitive behavior) that this task needs internal state, but the embodied examples shows that with only inertia (without internal state) it is also possible (I have experiments showing that already).

The second example is a bit more complex, it involves trying to get a reactive agent to solve the exact same problem of Beer's more recent (2003) paper in adaptive behaviour (The dynamics of active categorical perception in an evolved model agent). I am almost sure it could be possible that an agent with a purely reactive network can solve the task when the circle and the diamond are of fixed sizes (this is the same paper which you replicated while varying the sizes of the circles/diamonds, in which case I would not think could be solved with a reactive controller). This has two implications, the first one would be along the lines that we have been talking about, the importance of embodiment and situatedness. But the second one is a little bit different, it prompts the question, is the methodology that we are using all right? what are missing? when Beer evolves this agent to categorise between circles and diamonds (of fixed size), there is no pressure to evolve the simplest solution. And if the interest is in the analysis of the resulting dynamical system, as is the case in that paper (and is the case for my motivation), then, one ends up analysing something that is much more complex than could be needed. So, if it is the case that a reactive agent evolves to solve the task, placing the explanations on the agent/environment's historical interaction and not on the particular activations of the neurons, then the dynamical systems explanation that Randy offers is of less use. The particular suggestion to change our approach would be to have a cost associated to complexity in our methodology of evolving agents, time-constants should tend to be as small as possible by default (as reactive as possible), having slower acting neurons being costly.. , similar for other parameters such as connections, which may need to be costly as well, so networks tend to be as small as possible.. etc.

Anyway, new but related ideas and discussions are ongoing with Ezequiel which may end up in a journal publication (fingers crossed)...

3 Comments:

At 3:36 pm, Blogger eduardo said...

Interesting points made by other people with regards to my paper from the main conference are: [a] not agreeing with the definition of reactive behaviour (made by eldan) and [b] not agreeing with the title as it refers to embodiment as opposed to situatedness (point made by xabier).

 
At 3:41 pm, Blogger eduardo said...

oh. regarding replicating a reactive circle/diamond categoriser. i do not mean to say that such a result would invalidate the main points that Beer tries to say in his paper which can be seen as: [a] the importance of studying these systems dynamically and [b] looking at the simplest cognitive tasks possible (as pointed out to me by ezequiel). but rather that it would help contribute to the methodology of a dynamical systems approach to cognition by putting pressure on analysing as much as possible the simplest agents that can solve any such task.

 
At 5:31 pm, Blogger eduardo said...

While not invalidating any of Beer's claims, having a reactive circle-diamond discriminator would help in two major ways: (a) help put yet more emphasis into the agent-environment interaction in the dynamical systems understanding of cognition and (b) help revise the methodology to produce the simplest agents possible in addition to what we already do in ER which is to choose the simplest 'representationally-hungry' task possible (e.g. minimally cognitive behaviours), the simplest (i.e. unbiased) 'design' method (e.g. artificial evolution) and the simplest (i.e. lowest level possible) building blocks (i.e. CTRNNs).

 

Post a Comment

<< Home