"yan king yin" <y.k.y@(dont spam)lycos.com> wrote in message news:<Te578.7202$Lv.819466 at news.xtra.co.nz>...
> Im just reading an intro to SOMs and am wondering how could you construct
> a memory system out of one or several SOMs. I think any iterative mechanism
> is unlikely to occur in the brain, unless you can point out a specific circuit for it.
> Also the memory system has to be unsupervised. The SOM algorithm seems
> to require supervised training which makes it not so biologically plausible.
>
Not all neural networks are supervised. Learning to match input to a
specific output is one way of training a NN but there are many
paradigms that do not use this. I haven't read much on NNs (or SOMs)
but as far as i can tell 'supervised' and 'unsupervised' do not really
hold for SOMs. You start of with initial values for all nodes and
then as training data is input the nodes 'compete' for the data and
are modified by it. There is no specific map that is to be achieved,
the success of the SOM is judged by whether it can correctly classify
a set of test data. In a 'memory' system the training data would just
be the random sensory info the brain is exposed to, the test data
would be a retrieval operation on that data (of course we're taliing
about relatively 'simple' SOMs here, to test a map by 'questioning' it
would require a more sophisticated system)
As to iteration, there are many feedback mechanisms evident in the
nervous system and neuroscientists such as Edelman have posited
're-entry' as the fundamental mechanism by which consciousness arises.
I certainly don't know of any such circuit though. (if I did I'd
probably have a nobel prize lol)
> ------------------------- Original Message -----------------------------
> mat wrote:
>> The memory functions of the brain are certainly impressive but not of
> course not infallible. An interesting point is that we often seem to
> know when we have recollected a name or a fact (for example)
> incorrectly. By implication this must mean we either have the correct
> memory somewhere with which to compare it (in which case why don't we
> just recall that?) or that there is another error-checking system of
> some sort. Just wondering if anyone had any ideas.
>> One possible mechanism through which you get some of these properties
> would be to use two or more neural network-type systems (particularly
> the self-organizing map (SOM) paradigm). When a memory was to be
> accessed then the 'request' would be sent through the first SOM. Some
> output from this frist SOM would then be input to a second SOM. If
> the output from the second and first matched then the memory would be
> considered 'correct'. However, if not then the output of the second
> SOM would again be input to the first SOM. Eventually you would end
> up such that the output and input to each SOM was the same and this
> would then be considered correct (of course it still might actually be
> incorrect). If the outputs and inputs never converged then you would
> have 'forgotten'. The process is similar to the iterative formulas
> for finding equation roots where you use the output as subsequent
> input until there is no change (to a defined degree of accuracy)