Mailing List Archive

Mailing List: techdiver

Message Display

"Terry McCracken" <terrym@it*.ns*.co*.au*> writes:

>> There has been some discussion on the limitations of the dive
>> computers not having enough grunt to perform complex caclulations for
>> advanced dive alogrythims (close enough?). Has there been any research into
>> either some form of neural-nets type computation...

Then <devon@ol*.ch*> writes:

> I do research with a technique called stochastic discrimination which is
> a fast, low memory, general technique for solving problems like pattern
> recognition and optimization. I would love to tackle this problem and I've
> been keeping my eyes open for good data but I haven't run across any.

I have a background in neural networks and other soft-computing methods.
I think that there would be a verification problem with using NNs in dive
computers, and also that using NNs in this type of problem would be very
computationally intensive (although computing power is becoming cheaper
by the minute, and there are some good dedicated DSPs & NN processors
that could do the job).  It could certainly be done, but I don't see the
advantage that it would give you over the current approaches used in
dive computers.

Christina

Navigate by Author: [Previous] [Next] [Author Search Index]
Navigate by Subject: [Previous] [Next] [Subject Search Index]

[Send Reply] [Send Message with New Topic]

[Search Selection] [Mailing List Home] [Home]