Paris

===
Ayara et al. (2002) "Negative Selection: How to Generate Detectors." 

   -  negative selection metaphor for immune inspired fault tolerance.
   -  self non-self descrimination by negative selection
   -  self => (e.g.) normal network traffic between computers

Exhaustive detector generating algorithm (Forrest, Perersen et al 1994)

candidate non-self detector -> test "If match?" with self-data exhaustively

experiment with 8-bit binary pattern

N_self = 8 .. 160
N_non-self = 256 - N_self 

===
T. Yu & J. Miller (2002)
Finding Needles in Haystacks is Not Hard with Neutrality (EuroGP)

  vs

M. Collins (2004)
Finding Needles in Haystacks is Harder with Neutrality. 


Yu

even-12-parity more than 55 success from 100 runs each of a only max of 10,000 iterations
while with random search never succeeded in 4,000,000 trials.

1. Randomly generate an initial population of 5 genotypes with the lowest possible
fitness and select one (randomly) as the winner.
2. Carry out point-wise mutation on the winning parent to generate 4 offspring;
3. Construct a new generation with the winner and its offspring;
4. Select a winner from the current population using the following rules:

 If any offspring has a better fitness, it becomes the winner.

 Otherwise, an offspring with the same fitness is randomly selected. If the
parent-offspring pair has a Hamming distance within the permitted range
(see Section 6.2), the offspring becomes the winner.

 Otherwise, the parent remains as the winner.
5. Go to step 2 unless the maximum number of generations reached or a solution
with needle fitness is found.

(5-011)(6-251)(7-550)(8-630)(9-471)(10-841)

x0    x2
   EQ    EQ   xor      eq - y
x1          x3     x4


or

(1 eq a b 3)(2 eq c d 5)(3 eq 1 e out)
a eq b eq e

12-parity



=> one max problem
but no more needle in hay


=====
Collins "not-easy version"
an analysis of the reported successes
of the Cartesian Genetic Programming method on 
a simplified form of the Boolean parity problem. 

We present results indicating that 

the loss of performance is caused by the sampling 
bias of the CGP, due to the neu-trality friendly representation. 

We implement a simple in-tron free random sampling algorithm 
which performs con-siderably better on the same problem

 and then 
explain how such performance is possible.

We also implement and test a trivial random sampling
algorithm which uses a fully expressed genotype. 

We show
that this algorithm considerably exceeds expectations un-
expectedly performing better than random sampling. An
analysis of how simply removing introns from the represen-
tation can produce such a result is provided.





photo with j, k, me in minsk


colabaration including good student
frankly speaking, 
even seminor not so easy 
autority (read recture is enough)

project on the web
authority still not so interested in web
but seminor



classification => old yet still open

input n-d data output rule

Michigan vs Pits

turget island in lake

ultmate needle

fuzzy 
needle is crisp


supervised

train by self

cyrano


no-supervised