InterJournal Complex Systems, 291
Status: Submitted
Manuscript Number: [291]
Submission Date: 990701
Revised On: 990701
Referee Report for BArticle 247
Author(s): Anonymous

Subject(s): CX.67, CX.65, CX.31, CX.33, CX.35, CX.16, CX.13

Category: Brief Article

Abstract:

This article introduces several interesting ideas to core-memory-based genetic programming systems such as a flexible memory space using logical addressing rather than physical one, hierarchical partitioning of memory space by membranes, parallel execution of programming codes, etc., which would merit publication. The paper contains, however, some defects mainly in presentation as shown below.

The most serious problem is that what is novel in this Semar system compared to other evolutionary systems is unclear. The author utilizes only ordinary GAs to evolve creatures in Semar, and the touchstone problem tested in this paper is not more than a simple toy-like one. The author should place more emphasis on what is the feature of this system newly introduced in this study and what is derived from it as a new knowledge.

It is also a problem that the details of how virtual substances in Semar function are almost omitted in the paper so that readers can hardly imagine its workings. Definitions of several technical terms such as Type, Label, Value, Mnemonic, Address, and Template, are not obvious. More formal explanation is also needed to the structure and the function of Pure data, Membrane, Instruction, and Operator.

Readers may be interested in the application of this kind of genetic programming models. It is to be desired that the author provide a brief remark about this issue.

Detailed comments:

Chapter 1, Paragraph 2, middle:
Delete either 'self-' or 'themselves' in the phrase "as a soup in which program creatures self-replicate themselves."

Chapter 2, Figure 1.1 and 1.2; Chapter 3, Figure 1.3:
These figures are all blurry and unreadable. They should be replaced with much clearer ones.

Chapter 4:
This chapter should be entitled 'Summary' or something like that according to its contents, instead of 'Discussion'.

Retrieve Previous Revision's Abstract
Submit referee report/comment


Public Comments: