Learning to schedule job-shop problems: representation and policy learning using graph neural network and reinforcement learning

Junyoung Park, Junyoung Park, Jaehyeong Chun, Sang Hun Kim, Youngkook Kim, Jinkyoo Park, Jinkyoo Park · International Journal of Production Research · 2021 · machine-learning · 10.1080/00207543.2020.1870013 · record park2021-learning-to-schedule-job-shop-prob
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. discipline is a boundary case: the venue and problem are operations-research/production-research, the contribution is a machine-learning architecture and training scheme; I tagged 'machine-learning' because the novelty is the GNN+RL representation-and-policy learning, and the empirical section is a benchmark comparison. method 'system-or-tool-building' covers both the framework and the released pyjssp simulator; the evaluation is a benchmark/simulation study, which the single-value facet cannot express. The paper never claims optimality (relative error 12.6-20.9% against known optima/bounds), so replication status is 'untested'. PDF extraction is heavily garbled in the equation regions (equations 5-12 and the arXiv stamp on page 1); all quotes were taken from clean prose paragraphs and verified as exact substrings. 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.

Some quotes on this page contain (cid:NN). The PDF this text was extracted from encodes 5 characters in a way that names no letter, so the extractor could not say what they were. The quotes keep them because a quote is reproduced exactly or it proves nothing — the marks are the source’s, not the tagging’s.

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
standard deviation 3 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 4 concept tags on this paper, all 4 are grounded in a quoted sentence.

job shop scheduling problem MDP makespan priority dispatching rules Because the embedding layer works through pairwise interaction functions that are shared across all node pairs rather than tied to a fixed graph size, the same trained network can be run on graphs with different node counts and edge patterns, which is what carries the policy to unseen instance sizes Training on a graph representation forces the policy to key on the structural regularities common to all job shops instead of memorising one instance, which is why it transfers with no further training The learned policy beats fixed rules because it re-weights which state features matter at each decision, whereas a dispatching rule always consults the same predetermined features The dense waiting-job reward is used instead of makespan because a reward delivered only at episode end leaves the algorithm unable to assign credit to individual dispatching actions job shop scheduling problem (JSSP) priority dispatching rule graph state for the JSSP non-trivial state primitive transition graph neural network (GNN) number of jobs number of machines relative scheduling error computation time additional training discount factor learning rate GAE parameter clipping parameter value function coefficient entropy bonus coefficient regression (polynomial curve fitting) the fitted second-order polynomial curve coefficient of determination standard deviation mean and standard deviation error bars (one standard deviation) one standard deviation of errors mean average relative error 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://doi.org/10.1080/00207543.2020.1870013. The reading view exists for the papers whose licence permits it.

Every tag on this page

claim_relations 2 · claim_type 1 · concept_edges 1 · concepts 4 · content_tags 41 · controls 1 · limitations 10 · mechanisms 4 · 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 1 relation 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	job-shop scheduling 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: Because the embedding layer works through pairwise interaction functions that are shared across all node pairs rather than tied to a fixed graph size, the same trained network can be run on graphs with different node counts and edge patterns, which is what carries the policy to unseen instance sizes	`mechanisms[0]`		the tagger'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: Training on a graph representation forces the policy to key on the structural regularities common to all job shops instead of memorising one instance, which is why it transfers with no further training	`mechanisms[1]`		the tagger'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: The learned policy beats fixed rules because it re-weights which state features matter at each decision, whereas a dispatching rule always consults the same predetermined features	`mechanisms[2]`		the tagger'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: The dense waiting-job reward is used instead of makespan because a reward delivered only at episode end leaves the algorithm unable to assign credit to individual dispatching actions	`mechanisms[3]`		the tagger's words
limitations	Computation time for node embeddings grows rapidly (quadratically in the number of jobs) with instance size. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the practical scalability claim, not the solution quality: THE COMPUTATIONAL TIME TO COMPUTE THE NODE EMBEDDINGS RAPIDLY INCREASES with instance size	`limitations[0]`		the tagger's words
limitations	The MDP formulation omits practical shop-floor constraints such as batching, waiting-time regulations and flexible setups. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any shop-floor deployment reading: THE CURRENT FORMULATION DOES NOT CONSIDER MORE PRACTICAL CONSTRAINTS SUCH AS BATCH PROCESSES AND WAITING TIME REGULATIONS AND FLEXIBLE SHOP SETUP	`limitations[1]`		the tagger's words
limitations	The distribution used to generate training and test instances does not cover real-world job shops. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any claim of applicability to real job shops: THE DISTRIBUTION P0(g0) USED TO GENERATE THE TRAINING AND TEST JSSP INSTANCES IS NOT GENERAL ENOUGH TO COVER REAL-WORLD JSSPs	`limitations[2]`		the tagger's words
limitations	The comparison with prior RL schedulers could only test limited generalisation, because those baselines require train and test instances of the same size. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the comparison against prior RL schedulers: ONLY A LIMITED GENERALIZATION TEST COULD BE CONDUCTED FOR THE CASE WHERE THE CONJUNCTIVE CONSTRAINTS AND PROCESSING TIMES ARE DIFFERENT FROM THE TRAINING AND TEST DATA SETS	`limitations[3]`		the tagger's words
limitations	The learned scheduler is slower at inference than simple priority dispatching rules. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: any deployment where scheduling latency matters: THE PROPOSED GNN-SCHEDULER REQUIRES MORE COMPUTATIONAL TIME THAN PDRs	`limitations[4]`		the tagger's words
limitations	The headline comparison is scoped to instances drawn from the training distribution, and the comparison against the other learned schedulers is scoped by a restriction on testing instances. 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 the GNN scheduler beats dispatching rules and other RL schedulers in general: the PDR comparison is on instances from the training distribution, and the RL comparison carries a stated restriction on testing instances	`limitations[5]`	1.0	the tagger's words
limitations	The method's cost grows with problem size by the authors' own analysis, which is the quantity a scheduling practitioner would care about. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: applying this scheduler to large shops: the authors state embedding time grows in proportion to n2 + nm in the numbers of jobs and machines	`limitations[6]`	1.0	the tagger's words
limitations	The reference against which solution quality is judged is another solver's output, not a proven optimum supplied independently. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the optimality gaps reported for the validation set: the reference minimum makespan is computed by the Google OR tool rather than supplied independently	`limitations[7]`	1.0	the tagger's words
limitations	Training stops on a convergence criterion measured on the validation set, so the reported model is the one that stopped improving on that set. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the trained policy as a fixed artefact: training was terminated when the validation-set makespan converged, so the model is defined by that stopping rule	`limitations[8]`	1.0	the tagger's words
limitations	The architecture search behind the reported hyperparameters is reported only as a negative result, with no numbers attached. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. what this limits: the reported architecture as tuned: deeper and thicker networks were tested and are reported not to have improved performance, without the comparison being shown	`limitations[9]`	1.0	the tagger's words
transfers	The size-agnostic transferability of graph neural networks, demonstrated in robot control, wind power estimation and weather prediction, is imported into production scheduling to make a scheduler that works on unseen shop sizes 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	Proximal Policy Optimization, a policy-gradient algorithm from the RL literature, is adopted to train the representation and dispatching modules of a job-shop scheduler end to end 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	the 6x6 instances were dropped from the performance evaluation because only one benchmark problem exists at that size Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. eliminates: that a mean over problem sizes is dominated by a size represented by a single instance · outcome (stated by the tagger, no quotation): held	`controls[0]`	0.95	the tagger's words
concepts	job shop scheduling problem Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: job shop scheduling problem	`concepts[0]`		the paper's words
concepts	MDP Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: MDP	`concepts[1]`		the paper's words
concepts	makespan Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: makespan	`concepts[2]`		the paper's words
concepts	priority dispatching rules Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: priority dispatching rules	`concepts[3]`		the paper's words
senses	generalization - AS THE ABSTRACT CLAIMS IT: a superb capability, asserted without a scope 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	generalization - AS THE CONCLUSION SCOPES IT: trained on only small JSSPs and producing high-quality schedules for large ones, with the PDR comparison run on the training distribution 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	scheduling quality - operationalised entirely as makespan: the maximum completion time among jobs, which is the single quantity the reward function is built to minimise 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	builds-on → Proximal Policy Optimization (Schulman et al., cited as [37]/[38]) 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	extends → MARL [10] and MDQN [24] RL-based JSSP schedulers 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
concept_edges	job shop scheduling problem —operationalised-as→ MDP 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 the span writes 'JSSP'; endpoint abbreviated. also in: job shop scheduling problem · MDP	`concept_edges[0]`		the tagger's words
content_tags	regression (polynomial curve fitting) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: regression (polynomial curve fitting)	`content_tags.statistics[0]`		the tagger's words
content_tags	coefficient of determination Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: coefficient of determination	`content_tags.statistics[1]`		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. also in: standard deviation	`content_tags.statistics[2]`		the tagger's words
content_tags	error bars (one standard deviation) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: error bars (one standard deviation)	`content_tags.statistics[3]`		the tagger's words
content_tags	mean Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: mean	`content_tags.statistics[4]`		the tagger's words
content_tags	On test instances the trained GNN scheduler reached 14.7% average relative error. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[0]`		the tagger's words
content_tags	On published benchmark sets the GNN scheduler gave average relative errors from 12.6% to 20.9%. 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	A scheduler trained only on small randomly generated instances transfers to much larger instances, since the gap between the generically trained GNN0 and the size-matched GNN1-3 was small. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[2]`		the tagger's words
content_tags	The GNN scheduler reliably produced high-quality solutions across instance types. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[3]`		the tagger's words
content_tags	Computation time grows roughly with the square of the number of jobs (quadratic fit R2 = 0.998). Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[4]`		the tagger's words
content_tags	Stacking more than three embedding layers bought no measurable scheduling improvement while costing extra computation, so the message-passing depth was fixed at three. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[5]`		the tagger's words
content_tags	The GNN scheduler needs more computation time than the priority dispatching rules. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[6]`		the tagger's words
content_tags	Fewer than 20 episodes per parameter update left the state-value estimate with large variance, while more than 20 added no performance, fixing the batch size at 20 episodes. Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.findings[7]`		the tagger's words
content_tags	job shop scheduling problem (JSSP) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: job shop scheduling problem (JSSP)	`content_tags.definitions[0]`		the tagger's words
content_tags	priority dispatching rule Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: priority dispatching rule	`content_tags.definitions[1]`		the tagger's words
content_tags	graph state for the JSSP Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: graph state for the JSSP	`content_tags.definitions[2]`		the tagger's words
content_tags	non-trivial state Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: non-trivial state	`content_tags.definitions[3]`		the tagger's words
content_tags	primitive transition Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: primitive transition	`content_tags.definitions[4]`		the tagger's words
content_tags	graph neural network (GNN) Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: graph neural network (GNN)	`content_tags.definitions[5]`		the tagger's words
content_tags	makespan Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: makespan	`content_tags.definitions[6]`		the tagger's words
content_tags	number of jobs Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: number of jobs	`content_tags.variables[0]`		the tagger's words
content_tags	number of machines Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: number of machines	`content_tags.variables[1]`		the tagger's words
content_tags	relative scheduling error Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: relative scheduling error	`content_tags.variables[3]`		the tagger's words
content_tags	computation time Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: computation time	`content_tags.variables[4]`		the tagger's words
content_tags	additional training Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: additional training	`content_tags.variables[5]`		the tagger's words
content_tags	discount factor Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. also in: discount factor	`content_tags.variables[7]`		the tagger's words
content_tags	learning rate Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: RULED, NOT PREVIOUSLY ADJUDICATED. This quote carries an unmapped font glyph and was not covered by the 2026-08-25 ruling, which reached only rows flagged with glyph damage. The same reasoning applies and the same decision follows: THE QUOTE IS NOT EDITED. The unmapped characters are the ones carrying the claim, and no code-to-character table is possible - in one sentence of this corpus the same code is '=' once and 'tau' twice. Repair needs the source PDF and is a re-staging job, costed and refused (0 of 114 appendable; a per-record re-stage fixes 21 and breaks 223). Three sibling rows were found ALREADY ruled under this same pass and are untouched. also in: learning rate	`content_tags.variables[8]`		the tagger's words
content_tags	GAE parameter Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: RULED, NOT PREVIOUSLY ADJUDICATED. This quote carries an unmapped font glyph and was not covered by the 2026-08-25 ruling, which reached only rows flagged with glyph damage. The same reasoning applies and the same decision follows: THE QUOTE IS NOT EDITED. The unmapped characters are the ones carrying the claim, and no code-to-character table is possible - in one sentence of this corpus the same code is '=' once and 'tau' twice. Repair needs the source PDF and is a re-staging job, costed and refused (0 of 114 appendable; a per-record re-stage fixes 21 and breaks 223). Three sibling rows were found ALREADY ruled under this same pass and are untouched. also in: GAE parameter	`content_tags.variables[9]`		the tagger's words
content_tags	clipping parameter Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: RULED, NOT PREVIOUSLY ADJUDICATED. This quote carries an unmapped font glyph and was not covered by the 2026-08-25 ruling, which reached only rows flagged with glyph damage. The same reasoning applies and the same decision follows: THE QUOTE IS NOT EDITED. The unmapped characters are the ones carrying the claim, and no code-to-character table is possible - in one sentence of this corpus the same code is '=' once and 'tau' twice. Repair needs the source PDF and is a re-staging job, costed and refused (0 of 114 appendable; a per-record re-stage fixes 21 and breaks 223). Three sibling rows were found ALREADY ruled under this same pass and are untouched. also in: clipping parameter	`content_tags.variables[10]`		the tagger's words
content_tags	value function coefficient Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: RULED, NOT PREVIOUSLY ADJUDICATED. This quote carries an unmapped font glyph and was not covered by the 2026-08-25 ruling, which reached only rows flagged with glyph damage. The same reasoning applies and the same decision follows: THE QUOTE IS NOT EDITED. The unmapped characters are the ones carrying the claim, and no code-to-character table is possible - in one sentence of this corpus the same code is '=' once and 'tau' twice. Repair needs the source PDF and is a re-staging job, costed and refused (0 of 114 appendable; a per-record re-stage fixes 21 and breaks 223). Three sibling rows were found ALREADY ruled under this same pass and are untouched. also in: value function coefficient	`content_tags.variables[11]`		the tagger's words
content_tags	entropy bonus coefficient Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above. about this quote: RULED, NOT PREVIOUSLY ADJUDICATED. This quote carries an unmapped font glyph and was not covered by the 2026-08-25 ruling, which reached only rows flagged with glyph damage. The same reasoning applies and the same decision follows: THE QUOTE IS NOT EDITED. The unmapped characters are the ones carrying the claim, and no code-to-character table is possible - in one sentence of this corpus the same code is '=' once and 'tau' twice. Repair needs the source PDF and is a re-staging job, costed and refused (0 of 114 appendable; a per-record re-stage fixes 21 and breaks 223). Three sibling rows were found ALREADY ruled under this same pass and are untouched. also in: entropy bonus coefficient	`content_tags.variables[12]`		the tagger's words
content_tags	14.7% — average relative makespan error of the trained scheduler on the 100 in-distribution test instances Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[0]`		the tagger's words
content_tags	over 20% — average relative error of every dispatching rule except MOR on the same test set Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[1]`		the tagger's words
content_tags	12.6% to 20.9% — range of the scheduler's average relative error across the published benchmark families Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[2]`		the tagger's words
content_tags	4.4% and 2.7% — the two cases where the scheduler was second-best by instance size (20x30 and 20x50), and by how much it trailed the best rule Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[3]`		the tagger's words
content_tags	R2 = 0.9984 — fit quality of the quadratic model relating runtime to the number of jobs Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[4]`		the tagger's words
content_tags	R2 = 0.9815 — fit quality of the linear model relating runtime to the number of machines Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[5]`		the tagger's words
content_tags	K = 3 — number of stacked embedding layers (hops of message passing) used in the GNN Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[6]`		the tagger's words
content_tags	100 — number of held-out test instances sampled from the training distribution Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[7]`		the tagger's words
content_tags	20 machines and 100 jobs — size of the largest instance the scheduler was tested on Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[8]`		the tagger's words
content_tags	20 — episodes collected per PPO parameter update Quotation withheld — this record is not cleared for redistribution. The sentence is in the paper at the link above.	`content_tags.quantities[9]`		the tagger's words

Interpretive choices

Tags that could not be placed in the text:

none — every tag was located in the text

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.