Minibatching Offers Improved Generalization Performance for Second Order Optimizers Article Swipe
Eric A. Silk
,
Swarnita Chakraborty
,
Nairanjana Dasgupta
,
Anand D. Sarwate
,
Andrew Lumsdaine
,
Tony Chiang
·
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2307.11684
YOU?
·
· 2023
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2307.11684
Training deep neural networks (DNNs) used in modern machine learning is computationally expensive. Machine learning scientists, therefore, rely on stochastic first-order methods for training, coupled with significant hand-tuning, to obtain good performance. To better understand performance variability of different stochastic algorithms, including second-order methods, we conduct an empirical study that treats performance as a response variable across multiple training sessions of the same model. Using 2-factor Analysis of Variance (ANOVA) with interactions, we show that batch size used during training has a statistically significant effect on the peak accuracy of the methods, and that full batch largely performed the worst. In addition, we found that second-order optimizers (SOOs) generally exhibited significantly lower variance at specific batch sizes, suggesting they may require less hyperparameter tuning, leading to a reduced overall time to solution for model training.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2307.11684
- https://arxiv.org/pdf/2307.11684
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4385227318
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4385227318Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2307.11684Digital Object Identifier
- Title
-
Minibatching Offers Improved Generalization Performance for Second Order OptimizersWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2023Year of publication
- Publication date
-
2023-05-26Full publication date if available
- Authors
-
Eric A. Silk, Swarnita Chakraborty, Nairanjana Dasgupta, Anand D. Sarwate, Andrew Lumsdaine, Tony ChiangList of authors in order
- Landing page
-
https://arxiv.org/abs/2307.11684Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2307.11684Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2307.11684Direct OA link when available
- Concepts
-
Hyperparameter, Variance (accounting), Generalization, Computer science, Machine learning, Artificial neural network, Artificial intelligence, Order (exchange), Variable (mathematics), Mathematics, Accounting, Finance, Economics, Mathematical analysis, BusinessTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4385227318 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2307.11684 |
| ids.doi | https://doi.org/10.48550/arxiv.2307.11684 |
| ids.openalex | https://openalex.org/W4385227318 |
| fwci | |
| type | preprint |
| title | Minibatching Offers Improved Generalization Performance for Second Order Optimizers |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10320 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9958999752998352 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1702 |
| topics[0].subfield.display_name | Artificial Intelligence |
| topics[0].display_name | Neural Networks and Applications |
| topics[1].id | https://openalex.org/T10848 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9944999814033508 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1703 |
| topics[1].subfield.display_name | Computational Theory and Mathematics |
| topics[1].display_name | Advanced Multi-Objective Optimization Algorithms |
| topics[2].id | https://openalex.org/T12814 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9941999912261963 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1702 |
| topics[2].subfield.display_name | Artificial Intelligence |
| topics[2].display_name | Gaussian Processes and Bayesian Inference |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C8642999 |
| concepts[0].level | 2 |
| concepts[0].score | 0.8664467334747314 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q4171168 |
| concepts[0].display_name | Hyperparameter |
| concepts[1].id | https://openalex.org/C196083921 |
| concepts[1].level | 2 |
| concepts[1].score | 0.7105671167373657 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q7915758 |
| concepts[1].display_name | Variance (accounting) |
| concepts[2].id | https://openalex.org/C177148314 |
| concepts[2].level | 2 |
| concepts[2].score | 0.6996934413909912 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q170084 |
| concepts[2].display_name | Generalization |
| concepts[3].id | https://openalex.org/C41008148 |
| concepts[3].level | 0 |
| concepts[3].score | 0.6292185187339783 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[3].display_name | Computer science |
| concepts[4].id | https://openalex.org/C119857082 |
| concepts[4].level | 1 |
| concepts[4].score | 0.5604103207588196 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[4].display_name | Machine learning |
| concepts[5].id | https://openalex.org/C50644808 |
| concepts[5].level | 2 |
| concepts[5].score | 0.5354011058807373 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q192776 |
| concepts[5].display_name | Artificial neural network |
| concepts[6].id | https://openalex.org/C154945302 |
| concepts[6].level | 1 |
| concepts[6].score | 0.5091258883476257 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[6].display_name | Artificial intelligence |
| concepts[7].id | https://openalex.org/C182306322 |
| concepts[7].level | 2 |
| concepts[7].score | 0.5034119486808777 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q1779371 |
| concepts[7].display_name | Order (exchange) |
| concepts[8].id | https://openalex.org/C182365436 |
| concepts[8].level | 2 |
| concepts[8].score | 0.46850278973579407 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q50701 |
| concepts[8].display_name | Variable (mathematics) |
| concepts[9].id | https://openalex.org/C33923547 |
| concepts[9].level | 0 |
| concepts[9].score | 0.2516094446182251 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[9].display_name | Mathematics |
| concepts[10].id | https://openalex.org/C121955636 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q4116214 |
| concepts[10].display_name | Accounting |
| concepts[11].id | https://openalex.org/C10138342 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q43015 |
| concepts[11].display_name | Finance |
| concepts[12].id | https://openalex.org/C162324750 |
| concepts[12].level | 0 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q8134 |
| concepts[12].display_name | Economics |
| concepts[13].id | https://openalex.org/C134306372 |
| concepts[13].level | 1 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q7754 |
| concepts[13].display_name | Mathematical analysis |
| concepts[14].id | https://openalex.org/C144133560 |
| concepts[14].level | 0 |
| concepts[14].score | 0.0 |
| concepts[14].wikidata | https://www.wikidata.org/wiki/Q4830453 |
| concepts[14].display_name | Business |
| keywords[0].id | https://openalex.org/keywords/hyperparameter |
| keywords[0].score | 0.8664467334747314 |
| keywords[0].display_name | Hyperparameter |
| keywords[1].id | https://openalex.org/keywords/variance |
| keywords[1].score | 0.7105671167373657 |
| keywords[1].display_name | Variance (accounting) |
| keywords[2].id | https://openalex.org/keywords/generalization |
| keywords[2].score | 0.6996934413909912 |
| keywords[2].display_name | Generalization |
| keywords[3].id | https://openalex.org/keywords/computer-science |
| keywords[3].score | 0.6292185187339783 |
| keywords[3].display_name | Computer science |
| keywords[4].id | https://openalex.org/keywords/machine-learning |
| keywords[4].score | 0.5604103207588196 |
| keywords[4].display_name | Machine learning |
| keywords[5].id | https://openalex.org/keywords/artificial-neural-network |
| keywords[5].score | 0.5354011058807373 |
| keywords[5].display_name | Artificial neural network |
| keywords[6].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[6].score | 0.5091258883476257 |
| keywords[6].display_name | Artificial intelligence |
| keywords[7].id | https://openalex.org/keywords/order |
| keywords[7].score | 0.5034119486808777 |
| keywords[7].display_name | Order (exchange) |
| keywords[8].id | https://openalex.org/keywords/variable |
| keywords[8].score | 0.46850278973579407 |
| keywords[8].display_name | Variable (mathematics) |
| keywords[9].id | https://openalex.org/keywords/mathematics |
| keywords[9].score | 0.2516094446182251 |
| keywords[9].display_name | Mathematics |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2307.11684 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2307.11684 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2307.11684 |
| locations[1].id | doi:10.48550/arxiv.2307.11684 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2307.11684 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5083373508 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Eric A. Silk |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Silk, Eric |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5113002383 |
| authorships[1].author.orcid | |
| authorships[1].author.display_name | Swarnita Chakraborty |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Chakraborty, Swarnita |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5038566014 |
| authorships[2].author.orcid | https://orcid.org/0000-0003-4009-6281 |
| authorships[2].author.display_name | Nairanjana Dasgupta |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Dasgupta, Nairanjana |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5087162046 |
| authorships[3].author.orcid | https://orcid.org/0000-0001-6123-5282 |
| authorships[3].author.display_name | Anand D. Sarwate |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Sarwate, Anand D. |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5074260102 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-9153-6622 |
| authorships[4].author.display_name | Andrew Lumsdaine |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Lumsdaine, Andrew |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5088679899 |
| authorships[5].author.orcid | https://orcid.org/0000-0002-9817-2526 |
| authorships[5].author.display_name | Tony Chiang |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Chiang, Tony |
| authorships[5].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2307.11684 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2023-07-25T00:00:00 |
| display_name | Minibatching Offers Improved Generalization Performance for Second Order Optimizers |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10320 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9958999752998352 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1702 |
| primary_topic.subfield.display_name | Artificial Intelligence |
| primary_topic.display_name | Neural Networks and Applications |
| related_works | https://openalex.org/W2140186469, https://openalex.org/W4390421286, https://openalex.org/W4280563792, https://openalex.org/W4389724018, https://openalex.org/W4318719684, https://openalex.org/W3183136280, https://openalex.org/W4318559728, https://openalex.org/W2775233965, https://openalex.org/W2038157384, https://openalex.org/W2585279543 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2307.11684 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2307.11684 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2307.11684 |
| primary_location.id | pmh:oai:arXiv.org:2307.11684 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2307.11684 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2307.11684 |
| publication_date | 2023-05-26 |
| publication_year | 2023 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 53, 81, 126 |
| abstract_inverted_index.In | 100 |
| abstract_inverted_index.To | 32 |
| abstract_inverted_index.an | 46 |
| abstract_inverted_index.as | 52 |
| abstract_inverted_index.at | 113 |
| abstract_inverted_index.in | 6 |
| abstract_inverted_index.is | 10 |
| abstract_inverted_index.of | 37, 60, 67, 89 |
| abstract_inverted_index.on | 18, 85 |
| abstract_inverted_index.to | 28, 125, 130 |
| abstract_inverted_index.we | 44, 72, 102 |
| abstract_inverted_index.and | 92 |
| abstract_inverted_index.for | 22, 132 |
| abstract_inverted_index.has | 80 |
| abstract_inverted_index.may | 119 |
| abstract_inverted_index.the | 61, 86, 90, 98 |
| abstract_inverted_index.deep | 1 |
| abstract_inverted_index.full | 94 |
| abstract_inverted_index.good | 30 |
| abstract_inverted_index.less | 121 |
| abstract_inverted_index.peak | 87 |
| abstract_inverted_index.rely | 17 |
| abstract_inverted_index.same | 62 |
| abstract_inverted_index.show | 73 |
| abstract_inverted_index.size | 76 |
| abstract_inverted_index.that | 49, 74, 93, 104 |
| abstract_inverted_index.they | 118 |
| abstract_inverted_index.time | 129 |
| abstract_inverted_index.used | 5, 77 |
| abstract_inverted_index.with | 25, 70 |
| abstract_inverted_index.Using | 64 |
| abstract_inverted_index.batch | 75, 95, 115 |
| abstract_inverted_index.found | 103 |
| abstract_inverted_index.lower | 111 |
| abstract_inverted_index.model | 133 |
| abstract_inverted_index.study | 48 |
| abstract_inverted_index.(DNNs) | 4 |
| abstract_inverted_index.(SOOs) | 107 |
| abstract_inverted_index.across | 56 |
| abstract_inverted_index.better | 33 |
| abstract_inverted_index.during | 78 |
| abstract_inverted_index.effect | 84 |
| abstract_inverted_index.model. | 63 |
| abstract_inverted_index.modern | 7 |
| abstract_inverted_index.neural | 2 |
| abstract_inverted_index.obtain | 29 |
| abstract_inverted_index.sizes, | 116 |
| abstract_inverted_index.treats | 50 |
| abstract_inverted_index.worst. | 99 |
| abstract_inverted_index.(ANOVA) | 69 |
| abstract_inverted_index.Machine | 13 |
| abstract_inverted_index.conduct | 45 |
| abstract_inverted_index.coupled | 24 |
| abstract_inverted_index.largely | 96 |
| abstract_inverted_index.leading | 124 |
| abstract_inverted_index.machine | 8 |
| abstract_inverted_index.methods | 21 |
| abstract_inverted_index.overall | 128 |
| abstract_inverted_index.reduced | 127 |
| abstract_inverted_index.require | 120 |
| abstract_inverted_index.tuning, | 123 |
| abstract_inverted_index.2-factor | 65 |
| abstract_inverted_index.Analysis | 66 |
| abstract_inverted_index.Training | 0 |
| abstract_inverted_index.Variance | 68 |
| abstract_inverted_index.accuracy | 88 |
| abstract_inverted_index.learning | 9, 14 |
| abstract_inverted_index.methods, | 43, 91 |
| abstract_inverted_index.multiple | 57 |
| abstract_inverted_index.networks | 3 |
| abstract_inverted_index.response | 54 |
| abstract_inverted_index.sessions | 59 |
| abstract_inverted_index.solution | 131 |
| abstract_inverted_index.specific | 114 |
| abstract_inverted_index.training | 58, 79 |
| abstract_inverted_index.variable | 55 |
| abstract_inverted_index.variance | 112 |
| abstract_inverted_index.addition, | 101 |
| abstract_inverted_index.different | 38 |
| abstract_inverted_index.empirical | 47 |
| abstract_inverted_index.exhibited | 109 |
| abstract_inverted_index.generally | 108 |
| abstract_inverted_index.including | 41 |
| abstract_inverted_index.performed | 97 |
| abstract_inverted_index.training, | 23 |
| abstract_inverted_index.training. | 134 |
| abstract_inverted_index.expensive. | 12 |
| abstract_inverted_index.optimizers | 106 |
| abstract_inverted_index.stochastic | 19, 39 |
| abstract_inverted_index.suggesting | 117 |
| abstract_inverted_index.therefore, | 16 |
| abstract_inverted_index.understand | 34 |
| abstract_inverted_index.algorithms, | 40 |
| abstract_inverted_index.first-order | 20 |
| abstract_inverted_index.performance | 35, 51 |
| abstract_inverted_index.scientists, | 15 |
| abstract_inverted_index.significant | 26, 83 |
| abstract_inverted_index.variability | 36 |
| abstract_inverted_index.hand-tuning, | 27 |
| abstract_inverted_index.performance. | 31 |
| abstract_inverted_index.second-order | 42, 105 |
| abstract_inverted_index.interactions, | 71 |
| abstract_inverted_index.significantly | 110 |
| abstract_inverted_index.statistically | 82 |
| abstract_inverted_index.hyperparameter | 122 |
| abstract_inverted_index.computationally | 11 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 6 |
| citation_normalized_percentile |