CAN: Creative Adversarial Networks, Generating Art by Learning About Styles and Deviating from Style Norms

Ahmed Elgammal, Bingchen Liu, Mohamed Elhoseiny, Marian Mazzone · ICCC / arXiv · 2017 · art-history · record elgammal2017-creative-gan
tag layer · index · concepts · authors · עברית

Where this record's text came from. faceted 2026-08 by reading the locally held text (pdftext); every evidence string verified as an exact substring of that text before merging. The paper is a hybrid: a built system plus four human-subject validation studies ('Quotation withheld by artists.'). I coded method as system-or-tool-building because the system is the contribution and the experiments are validation, but behavioural-experiment is a defensible second method tag. The paper's critique of plain GANs ('We argue that such networks are limited in their ability to generate creative products in their original design.') is an argued limitation rather than an empirical contradiction, so it is recorded as the 'extends' relation rather than 'contradicts'. Experiment III reused Snapper et al's instrument but on different stimuli, so it is coded builds-on, not replicates. 2026-08-14 and 2026-08-25 schema-0.3 passes: every quote verified as an exact substring of the staged text and anchored at merge time; concept terms verified present; every edge endpoint an existing node. Update, 15 September 2026: EXCEPTION, measured with the merge gate's own predicate (Paper.locate): 1 quote(s) on this record are not found as one exact string in the current staged text - content_tags.findings[0].evidence. The sentence above does not hold for them as written. Where a row records why - the quote is grounded in another copy of the paper, joins two passages with an ellipsis, or differs from the staged text by extraction noise - that note stays on the row.

The quality of the text these quotes were cut from. symbol-font-not-extractable: 5 '(cid:NNN)' glyph-index markers remain in a staged text of 49,975 characters, where the PDF's embedded font had no ToUnicode map. THE QUOTES ON THIS RECORD ARE THE PAPER'S AND THE STAGED COPY IS THE DAMAGED ONE - which is the opposite of the usual case and is why it went unnoticed: one of its 51 live quotes is not an exact substring of this text, and it could NOT be located in it even by aligning the quote's WORD SEQUENCE under this damage, so the defect recorded here is present on the record but does not by itself account for the miss. Not repaired in place: editing the staged text moves every offset after the edit and re-hashes every anchor on the record. Measured at 113aae60fefc3eb1, re-measured unchanged at 54b63482778ddf56.

Some words in these quotes contain the wrong character. The PDF this text was extracted from substituted 2 characters inside a word — a currency sign or a stray accent where a letter pair belongs, as in signi¢cantly for significantly, or a single ligature glyph, as in conﬁdence. They are valid characters in the wrong place, so they survive every encoding check. The quotes keep them for the same reason as above, and a search of this page will not find the affected words.

2 quotes on this page break a word across a line. The PDF this text was extracted from hyphenated those words at the end of a printed line — adversar- ial, repre- sentations — and the extractor kept the hyphen and the space. The quotes keep them because a quote is reproduced exactly or it proves nothing: the break is the typesetter’s, not the author’s. A search of this page will not find the affected words.

1 quote on this page was cut from a different copy of this article. This project holds two extractions of this paper: the plain text every other quote here is anchored in, and the publisher’s own HTML edition, which we are not licensed to republish. It is verbatim in the second and cannot be found in the first — not because the tagging is wrong, but because the extraction damaged the passage: a page number dropped into the middle of a number, a formula moved out of its sentence, a ligature where two letters belong. The record names the copy for each one. The quote is the paper’s; the extraction is not.

How to read this page. The article itself is not reproduced here — this paper is not cleared for redistribution, so what follows is the tag layer alone. The evidence sentences are withheld with the article. The table at the foot lists every tag the record carries; each row names the facet it belongs to, and the whose words column says whether the tag records the paper's own wording or the tagger's sentence about the paper. Every tag links to each other paper in the corpus carrying it. The machine-readable edition of the same thing is the TEI file beside this one, where the tags live in standOff, and the article is at its publisher via the link below. When we have changed our mind. Some entries in the table begin with a word in capitals and the reason for it. EVIDENCE CORRECTED — the finding still stands, but we had quoted the wrong sentence from the paper, and the quotation has been replaced with the right one. RETRACTED — the entry itself was wrong; it stays on the page, marked, with the reason, rather than being deleted. WITHDRAWN AND REPLACED — the entry was wrong, and a corrected version of it appears in the entries below. Each mark carries its own reason, in brackets, because the reasons differ from one entry to the next. An entry we have WITHDRAWN is removed from this page and kept in the record, with its reason. A line reading Audit and a date is a later re-reading of the entry against its whole sentence: DOUBTFUL means it was kept, and the reason says what is loose about it. Quotation marks inside the tagger's sentence. Where a tag is the tagger's own sentence, a phrase in quotation marks inside it is borrowed from the paper — the tagger's wording around it is not. The paper's own sentence is the evidence quote for that row; on this page it is withheld with the article, and the article is at its publisher. Nothing inside quotation marks anywhere on this page is ours: if it is quoted, it is the paper's. Whose words a concept carries. A concept entry records a term the paper itself uses, so the term, the sense given for it, and its evidence are the paper's own words, quoted where the paper fixes them. This is the opposite of a sense entry in the tag table, which is the tagger's sentence about how the paper uses a term. Where a concept entry could not be grounded in the paper's words, the entry says so on its own row.

Other fields define these differently the same word, defined another way in another discipline
art 5 fields · null-hypothesis 3 fields · novelty 3 fields · standard-deviation 3 fields · arousal 2 fields

Tags on this paper

Every tag the record carries, each linking to every other paper in the corpus that carries it. A tag appears here whether or not its words occur in the article — the spans below are only the ones a sentence in this paper licenses. Of the 8 concept tags on this paper, 5 are grounded in a quoted sentence.

creative adversarial network generative art style computational creativity style class style ambiguity collative variables novelty modify the GAN objective to push generated images to deviate from established style norms while staying near the art distribution art creative GAN arousal generated discriminator deviating Creative Adversarial Network (CAN) Wundt curve arousal potential hypothesis-testing null-hypothesis null hypothesis p-value standard-deviation standard deviation t-test system-or-tool-building

Legend

relation to other work	`claim_relations`	how this paper stands to another: agrees, contradicts, extends, qualifies
claim type	`claim_type`	the shape of the claim: existence, mechanism, method, critique, prediction, null
concept edge	`concept_edges`	an asserted relation between two concepts, with the sentence asserting it
concept	`concepts`	an idea the paper works with, quoted where the paper uses it
content tag	`content_tags`	a finding, definition, quantity, statistic or variable named in the paper
control	`controls`	an experiment run to eliminate a rival explanation
limitation	`limitations`	what the paper says its evidence cannot support
mechanism	`mechanisms`	how the reported effect is said to come about
method	`method`	how the paper produced its evidence
phenomenon	`phenomenon`	the effect or regularity under study
sense	`senses`	a distinct use of a term, quoted where the paper fixes it
transfer	`transfers`	a method or concept carried in from another field

The paper

The full text is not reproduced here. This article is not cleared for redistribution, so the page carries the tag layer only. Its evidence sentences are withheld; the grounding status remains with each tag, and the article itself is at https://arxiv.org/abs/1706.07068. The reading view exists for the papers whose licence permits it.

Every tag on this page

claim_relations 5 · claim_type 1 · concept_edges 2 · concepts 5 · content_tags 10 · controls 1 · limitations 9 · mechanisms 1 · method 1 · phenomenon 1 · senses 3 · transfers 2

Relations on this page carry a quotation for the claim. Where a relation was separately evidenced — and so tested against the vocabulary for its type — the row says so; none of 2 relations here carry that mark. The rest are the tagger’s reading of the sentence quoted beside them.

facet	tag	field	conf.	whose words
claim_type	method Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_type`		the paper's words
phenomenon	machine generation of creative visual art Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`phenomenon`		the paper's words
method	system-or-tool-building Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: system-or-tool-building	`method`		the paper's words
mechanisms	mechanisms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: modify the GAN objective to push generated images to deviate from established style norms while staying near the art distribution	`mechanisms[0]`		the tagger's words
limitations	GANs have a fundamental limitation in generating genuinely creative works Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the generative approach itself: Quotation withheld	`limitations[0]`		the tagger's words
limitations	Creativity of machine-generated artifacts is acknowledged to be an unresolved assessment problem. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: every creativity claim in the paper: ASSESSING THE CREATIVITY OF ARTIFACTS GENERATED BY THE MACHINE IS AN OPEN AND HARD QUESTION	`limitations[1]`		the tagger's words
limitations	Creativity was operationalised only as novelty, because influence could not be measured. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: what the word creative means in this work: SINCE INFLUENCE IS NOT RELEVANT HERE, WE USE NOVELTY AS A PROXY FOR CREATIVITY	`limitations[2]`		the tagger's words
limitations	The system has no semantic understanding of art beyond style categories. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any semantic reading of what the system does: IT DOES NOT HAVE ANY SEMANTIC UNDERSTANDING OF ART BEHIND THE CONCEPT OF STYLES	`limitations[3]`		the tagger's words
limitations	The authors cannot explain why subjects rated CAN images above the human Art Basel set. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the headline comparison, which the authors decline to interpret: WE LEAVE OPEN HOW TO INTERPRET THE HUMAN SUBJECTS' RESPONSES THAT RANKED THE CAN ART BETTER THAN THE ART BASEL SAMPLES	`limitations[4]`		the tagger's words
limitations	Further experiments are needed to settle the interpretation of the results. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the same scope as limitations[4]: MORE EXPERIMENTS ARE DEFINITELY NEEDED TO HELP ANSWER THESE QUESTIONS	`limitations[5]`		the tagger's words
limitations	Creativity is implemented as making a classifier maximally unsure which style it is looking at. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the creativity claimed for these images: it is implemented as maximising the entropy of the style class posterior, that is, as being hard for the style classifier to assign to any class	`limitations[6]`	1.0	the tagger's words
limitations	The style vocabulary is a fixed list whose length is a design parameter of the network. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the style categories used here: they are a fixed set whose size K is a network design parameter, fixed by the label set of the training corpus rather than by art history	`limitations[7]`	1.0	the tagger's words
limitations	The paper's own object is scare-quoted in its title. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the claim that this system generates art: the paper's title puts the word in quotation marks and describes the method as learning about styles and deviating from style norms	`limitations[8]`	1.0	the tagger's words
transfers	generative adversarial networks adapted to generate art that deviates from style norms Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`transfers[0]`		the tagger's words
transfers	Berlyne theory of aesthetic novelty and arousal used to design the creativity objective Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`transfers[1]`		the tagger's words
controls	an art-trained DCGAN baseline at both 64x64 and 256x256, generated and rated alongside the proposed model by the same human subjects Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: the human ratings reflecting that the images are art-like at all rather than that the proposed objective did anything - the baseline is the same architecture without it · outcome (stated by the tagger, no quotation): held	`controls[0]`	0.85	the tagger's words
concepts	style Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: style	`concepts[2]`		the paper's words
concepts	style class Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: style class	`concepts[4]`		the paper's words
concepts	style ambiguity Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: style ambiguity	`concepts[5]`		the paper's words
concepts	collative variables Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: collative variables	`concepts[6]`		the paper's words
concepts	novelty Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: novelty	`concepts[7]`		the paper's words
senses	style - AS A CLASS IN A CLASSIFIER: the network's final layer has K outputs where K is the number of style classes, so here the word names a position in a label set rather than any property of the picture, and an image's style is whichever index wins Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[0]`	1.0	the tagger's words
senses	style ambiguity - AS POSTERIOR ENTROPY: the generator is rewarded for making the style classifier uncertain, so 'deviating from style norms' is implemented as producing images the classifier cannot place, and creativity becomes a property of a confusion matrix Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[1]`	1.0	the tagger's words
senses	novelty - INHERITED FROM BERLYNE'S COLLATIVE VARIABLES: it is one of five arousal-raising properties named in an aesthetics theory - novelty, surprisingness, complexity, ambiguity and puzzlingness - and this system implements the fourth of them while the record's limitation says creativity was operationalised as the first Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`senses[2]`	1.0	the tagger's words
claim_relations	extends → Generative Adversarial Networks (from AI) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[0]`		the tagger's words
claim_relations	builds-on → Generative Adversarial Networks (GAN) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[1]`		the tagger's words
claim_relations	extends → Generative Adversarial Networks (GAN) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[2]`		the tagger's words
claim_relations	builds-on → Martindale's principle of least effort and explanation of style breaks Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[3]`		the tagger's words
claim_relations	builds-on → the art-perception experiment of Snapper et al [23] Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`claim_relations[4]`		the tagger's words
concept_edges	style ambiguity —operationalised-as→ style class Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. strength of the relation: 1 Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but ambiguity is achieved by maximizing the style-class posterior entropy; operationalised-as is loose. also in: style ambiguity · style class	`concept_edges[0]`		the tagger's words
concept_edges	novelty —part-of→ collative variables Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. strength of the relation: 1 also in: novelty · collative variables	`concept_edges[1]`		the tagger's words
content_tags	Experiments compared human subjects' responses to the generated art with their responses to art created by people. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[1]`		the tagger's words
content_tags	arousal potential Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: arousal potential	`content_tags.variables[0]`		the tagger's words
content_tags	Creative Adversarial Network (CAN) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: QUOTE REPAIRED: the extractor split a hyphen-space and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. also in: Creative Adversarial Network (CAN)	`content_tags.definitions[0]`		the tagger's words
content_tags	collative variables Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: collative variables	`content_tags.definitions[1]`		the tagger's words
content_tags	Wundt curve Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: Wundt curve	`content_tags.definitions[2]`		the tagger's words
content_tags	hypothesis-testing Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: QUOTE REPAIRED: the extractor split a unicode and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. also in: hypothesis-testing	`content_tags.statistics[0]`		the tagger's words
content_tags	null-hypothesis Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: QUOTE REPAIRED: the extractor split a unicode and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. also in: null-hypothesis	`content_tags.statistics[1]`		the tagger's words
content_tags	p-value Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: p-value	`content_tags.statistics[2]`		the tagger's words
content_tags	standard-deviation Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: QUOTE REPAIRED: the extractor split a hyphen-space and the quote was recorded faithfully with the damage in it, which made it unanchorable. The repaired string is an EXACT SUBSTRING of the staged text - that is why it was accepted, not because a rule produced it. The text as originally recorded is kept in evidence as extracted. also in: standard-deviation	`content_tags.statistics[3]`		the tagger's words
content_tags	t-test Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: t-test	`content_tags.statistics[4]`		the tagger's words

Interpretive choices

Tags that could not be placed in the text:

content_tags.findings[0] — quotation withheld — the page has already reproduced as much of this paper as the rights budget allows
NOT MACHINE-VOUCHED: declared-ellipsis - An elided quote: its 2 pieces ('The system generates art by look- ing at' ... 'izing de- viation from art distribution.') are each in our copy of the paper (with line-break hyphens, spacing and punctuation set aside), and the quote marks the cut with an ellipsis, so the joined string is in no copy. Located mechanically, piece by piece.

What was deliberately not tagged, and why, is recorded per paper by the tagger. Where a paper has no such record, this page says so rather than implying that nothing was declined.