A grand unified theory of everything

[Michael Oltmans <moltmans@>]

A grand unified theory of everything.

After centuries of research and thoughtful analysis of data from all 
fields, we here at GSB have developed a grand unifying theory of 
everything.  Contrary to common belief everything in every field can be 
unified under a common theme.  Our unifying theme consists of the 
following algorithm.

1) We commence with Bayes rule.

2) Actually, that was a joke, but at least Berthold won't waste his time.

3) Randomly select data from a randomly chosen field.

4) Smooth the data using convolutions of random mixtures of Gaussians 
with uniform randomly drawn mean and variance.

5) Randomly select a learning algorithm.

6) Train you learning algorithm.

7) Repeat steps 5-6 a random number of times to create a "panel of experts"

8) Randomly select a form of boosting algorithm and train it using the 
"panel of experts"

9) Test the algorithm on randomly sampled data from a randomly selected 
field, being careful to not use a field used for any part of the training.

10) This is your grand unifying theory of everything.

As evidence that this algorithm is infallible I have tested it on a 
series of questions after training on data from statistical physics and 
theology:

a) Where is all the fun?
b) Where can I find the meaning of life?
c) Where can I get a thesis topic that will get me a host of academic 
jobs to choose from?
d) Where is the site of the labs most holy weekly pilgrimage?
e) Where's the good stuff?

And the unanimous, and correct answer, to all of these questions was:

           +-                                                  -+
             girl scout benefit -+-  5:30 pm  -+- 32-G9 lounge
           +-                                                  -+

              For those coming from elsewhere: Building 32 is
               <http://whereis.mit.edu/map-jpg?selection=32>
          Once you are in 32, just take the G-elevator to the 9th
      floor and we will be in the lounge that you will be looking at
                    <http://projects.csail.mit.edu/gsb>












_______________________________________________
Csail-related mailing list
Csail-related@
https://lists.csail.mit.edu/mailman/listinfo/csail-related