By Lia Morra
Review Details
Reviewer has chosen not to be Anonymous
Overall Impression: Good
Content:
Technical Quality of the paper: Good
Originality of the paper: Yes
Adequacy of the bibliography: Yes
Presentation:
Adequacy of the abstract: Yes
Introduction: background and motivation: Bad
Organization of the paper: Satisfactory
Level of English: Unsatisfactory
Overall presentation: Average
Detailed Comments:
After carefully reviewing the revised manuscript, and taking into account reviewers’ responses, I believe that the manuscript was substantially improved. Nonetheless, I suggest a further round of revisions to further strengthen readability. The revised paper is also quite long and, while I do not have any objections in principle, a strong organization and clear message become even more important, to make the essential concepts stand out also to the less attentive readers.
- I would consider revising the introduction. It starts by defining the proposed approach as “unusual”, which I find a bizarre choice of wording. Then it defines the content of the paper section by section and then the contribution. My suggestion would be to write a more conventional introduction summarizing the main issue, the limitations of existing work and the main contribution, and optionally a paragraph illustrating how the rest of the paper is organized. Parallelly, I would move the context subsections to a separate section (Context or Related work) and move the notation to Section 2
- I am still not convinced by the definitions of symbols, icons and index in semiotics (page 5) – nor whether this introduction is essential to the rest of the paper. To the best of my knowledge, an “index” is not a special case of “icon”, but rather a different class of sign altogether (cfr. “The type of representational relation the sign bears to its object, or more precisely, to its ground, determines whether a sign is an Icon, and Index, or a Symbol.” [1]). In semiotics there is no notion of hierarchy, which as far as I understand is introduced by the authors in relation to the complexity mechanisms which, from computational neuro-science, could support the emergence of different types of signs as studied in semiotics. However, this argument is not made in the paper, and the correspondence between the indexical sign and the hippocampus is not clear to me as object features are not necessarily the only type of index available. Another interesting point to clarify is the relationship between the concept of index, icon and symbol from Figure 3 and the data structure presented in Figure 4.
- References to individual requests of the reviewers, including naming the reviewers, is highly unusual and I don't see the point. If the authors wish, generally thanking the reviewers in an Acknowledgement section is sufficient. The content of the paper is the authors’ responsibility, not the reviewers’. I would also use Acknowledgements instead of footnotes for contributions that do not meet authorship requirements (e.g., footnote 52)
- In some sections of the paper, many paragraphs contain a single sentence. This happens in the introduction, but also Section 2.2, Section 2.4 and other. I suggest revising as it breaks the flow.
- In Section 4.3 it is not clear what is the goal or final objective of the agent in the reinforcement learning example, and why the symbolic approach should be better a priori. Figure 15 is not referenced in the text. This example is quite useful and, while details can be found in the referenced report, the description of the task should remain here. It is also useful to report if the task is a standard benchmark from the literature and, in that case, how do the results compare against SOTA techniques.
- I suggest a further round of revision for typos and minor stylistic adjustments – a partial list is provided at the end. I would also move the references before the Appendix, and not after.
[1] Sánchez-Ovcharov, Carmen, and Mauricio Suárez. "Peirce’s Pragmatism, Semiotics, and Physical Representation." European Journal of Pragmatism and American Philosophy 16.XVI-1 (2024).
Minor and typos
- At page 2 line 11, “The notion of extrapolation, allowing exploration and search mechanisms is also designed”: what do you mean by designing extrapolation?
- At page 3 line 8 a priory instead of a priori
- At page 3 line 46 In the [63] it is also equipped with a rather efficient algorithm.: While I would in general avoid the use of citations as subjects or objects, if used then I would not use “the”. (In [63] it is also equipped with).
- Page 5 lines 9-15: please check for flow and proper use of citation quotes
- Page 6 line 41 broken reference
- Page 6 “Concepts are represented as a hierarchical data structure, in the sense of [23], how made the distinction between associative, sequential and hierarchical cognitive memories”. Typo or grammar issue (probably who instead of how?)
- Several references are broken, e.g., Page 6 line 41, caption of Figure 4, Page 7 line 40
- Page 7 lines 40 41: “in compliance compliant with the author formalization proposal”. Repetition – also, it is not clear who the author refers to
- Page 12 line 40 Contrary should not be capitalized
- Pay attention to the proper formatting of item list, e.g., see page 13 lines 1-3
- Section 2.6 starts with “Such a projector” but it is not clear to which projector it refers to
- Page 14 L1-2: List and set can either be empty by default, or equal to a list or set with default value. -> Lists and sets can either be empty by default…
- Ensure that all titles are numbered (e.g. Page 17 Line 12 is not)
- Page 17 Lines 16-19: probably itemization can be removed here
- Section 4 title could be rephrased as Illustrative examples or Application examples?
- Page 20 line 19 pay attention to the url embedded in the text, here and in the rest of the page, there are some issues with footnotes
- Section 5.1 rephrase title as “On the computational efficiency of the method”?
