REJECT: Article Is Not Appropriate For The IEEE Transactions On Evolutionary Computation
REJECT: Article Is So Poorly Written That Contributions, If Any, Are Obscure
REJECT: The Results Are Not Correct

Comments
 
-----------------------------------
Confidential Comments to the Editor
-----------------------------------
The topic is too old and not proposed a new method or new findings.
In additon, the paper is a patchwork from other papers.

----------------------
Comments to the Author
----------------------

(1)
If not a plagiarism, it's not good to use an expression as it is which you found in another paper.
I recommend you to write in such a case, like for example.

"1982, Hopfield [6] proposed a fully connected neural network model of associative memory in which we can store information by distributing it among neurons, and recall it from the neuron states dynamically relaxed AS IMADA DESCRIBED [23]."

The above is just one of many. The reviewer found many such expressions from other papers without mentioning it.

(2)
The topic is too old to be appeared nowadays. It might be good if it were in early 1990s when we stil had many discussions on different weight configurations which gave a network a function of associative memory other than Hebb's.

The author wrote "Perhaps the most common and the best performing general training algorithm currently used for Hopfield type networks is the pseudo-inverse algorithm [8]."

Can we say it "currently?" The paper [8] was published in 1986."

(3)
The reviewer doubts the result more or less. The author claims that fully connected 35 neurons store 26 patterns. This suggests storage capacity is about 74% of the number of neurons, while the Hopfield neural net work has at most 15% which was already analytically proved.

Even in the case that all the basins of attraction are very narrow as author wrote, this comparatively high storage capacity would be still quite interesting, though the authors do not mention about it.

(4)
The list of references is too biased and mostly those published more than a decade ago except for the authors' own publications, which means that survey on this topic is not sufficient.