Instance-Conditional Timescales of Decay for Non-Stationary Learning Article Swipe
Nishant Jain
,
Pradeep Shenoy
·
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.1609/aaai.v38i11.29173
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.1609/aaai.v38i11.29173
Slow concept drift is a ubiquitous, yet under-studied problem in practical machine learning systems. In such settings, although recent data is more indicative of future data, naively prioritizing recent instances runs the risk of losing valuable information from the past. We propose an optimization-driven approach towards balancing instance importance over large training windows. First, we model instance relevance using a mixture of multiple timescales of decay, allowing us to capture rich temporal trends. Second, we learn an auxiliary scorer model that recovers the appropriate mixture of timescales as a function of the instance itself. Finally, we propose a nested optimization objective for learning the scorer, by which it maximizes forward transfer for the learned model. Experiments on a large real-world dataset of 39M photos over a 9 year period show upto 15% relative gains in accuracy compared to other robust learning baselines. We replicate our gains on two collections of real-world datasets for non-stationary learning, and extend our work to continual learning settings where, too, we beat SOTA methods by large margins.
Related Topics
Metadata
- Type
- article
- Language
- en
- Landing Page
- http://doi.org/10.1609/aaai.v38i11.29173
- OA Status
- diamond
- Cited By
- 2
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4393153159
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4393153159Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.1609/aaai.v38i11.29173Digital Object Identifier
- Title
-
Instance-Conditional Timescales of Decay for Non-Stationary LearningWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-03-24Full publication date if available
- Authors
-
Nishant Jain, Pradeep ShenoyList of authors in order
- Landing page
-
https://doi.org/10.1609/aaai.v38i11.29173Publisher landing page
- Open access
-
YesWhether a free full text is available
- OA status
-
diamondOpen access status per OpenAlex
- OA URL
-
https://doi.org/10.1609/aaai.v38i11.29173Direct OA link when available
- Concepts
-
Statistical physics, Computer science, Physics, MathematicsTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
2Total citation count in OpenAlex
- Citations by year (recent)
-
2024: 2Per-year citation counts (last 5 years)
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4393153159 |
|---|---|
| doi | https://doi.org/10.1609/aaai.v38i11.29173 |
| ids.doi | https://doi.org/10.1609/aaai.v38i11.29173 |
| ids.openalex | https://openalex.org/W4393153159 |
| fwci | 2.24206349 |
| type | article |
| title | Instance-Conditional Timescales of Decay for Non-Stationary Learning |
| biblio.issue | 11 |
| biblio.volume | 38 |
| biblio.last_page | 12781 |
| biblio.first_page | 12773 |
| topics[0].id | https://openalex.org/T10502 |
| topics[0].field.id | https://openalex.org/fields/22 |
| topics[0].field.display_name | Engineering |
| topics[0].score | 0.9388999938964844 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/2208 |
| topics[0].subfield.display_name | Electrical and Electronic Engineering |
| topics[0].display_name | Advanced Memory and Neural Computing |
| topics[1].id | https://openalex.org/T11447 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9243000149726868 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1711 |
| topics[1].subfield.display_name | Signal Processing |
| topics[1].display_name | Blind Source Separation Techniques |
| topics[2].id | https://openalex.org/T10682 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9226999878883362 |
| 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 | Quantum Computing Algorithms and Architecture |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C121864883 |
| concepts[0].level | 1 |
| concepts[0].score | 0.493085116147995 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q677916 |
| concepts[0].display_name | Statistical physics |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.4001176655292511 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C121332964 |
| concepts[2].level | 0 |
| concepts[2].score | 0.3567556142807007 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q413 |
| concepts[2].display_name | Physics |
| concepts[3].id | https://openalex.org/C33923547 |
| concepts[3].level | 0 |
| concepts[3].score | 0.3426351547241211 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[3].display_name | Mathematics |
| keywords[0].id | https://openalex.org/keywords/statistical-physics |
| keywords[0].score | 0.493085116147995 |
| keywords[0].display_name | Statistical physics |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.4001176655292511 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/physics |
| keywords[2].score | 0.3567556142807007 |
| keywords[2].display_name | Physics |
| keywords[3].id | https://openalex.org/keywords/mathematics |
| keywords[3].score | 0.3426351547241211 |
| keywords[3].display_name | Mathematics |
| language | en |
| locations[0].id | doi:10.1609/aaai.v38i11.29173 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4210191458 |
| locations[0].source.issn | 2159-5399, 2374-3468 |
| locations[0].source.type | conference |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | 2159-5399 |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| locations[0].source.host_organization | https://openalex.org/P4310320058 |
| locations[0].source.host_organization_name | Association for the Advancement of Artificial Intelligence |
| locations[0].source.host_organization_lineage | https://openalex.org/P4310320058 |
| locations[0].source.host_organization_lineage_names | Association for the Advancement of Artificial Intelligence |
| locations[0].license | |
| locations[0].pdf_url | |
| locations[0].version | publishedVersion |
| locations[0].raw_type | journal-article |
| locations[0].license_id | |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| locations[0].landing_page_url | http://doi.org/10.1609/aaai.v38i11.29173 |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5073570615 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-3260-2543 |
| authorships[0].author.display_name | Nishant Jain |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Nishant Jain |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5110759892 |
| authorships[1].author.orcid | |
| authorships[1].author.display_name | Pradeep Shenoy |
| authorships[1].author_position | last |
| authorships[1].raw_author_name | Pradeep Shenoy |
| authorships[1].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | http://doi.org/10.1609/aaai.v38i11.29173 |
| open_access.oa_status | diamond |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Instance-Conditional Timescales of Decay for Non-Stationary Learning |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T03:46:38.306776 |
| primary_topic.id | https://openalex.org/T10502 |
| primary_topic.field.id | https://openalex.org/fields/22 |
| primary_topic.field.display_name | Engineering |
| primary_topic.score | 0.9388999938964844 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/2208 |
| primary_topic.subfield.display_name | Electrical and Electronic Engineering |
| primary_topic.display_name | Advanced Memory and Neural Computing |
| related_works | https://openalex.org/W2748952813, https://openalex.org/W2935759653, https://openalex.org/W3105167352, https://openalex.org/W54078636, https://openalex.org/W2954470139, https://openalex.org/W1501425562, https://openalex.org/W2902782467, https://openalex.org/W3084825885, https://openalex.org/W2298861036, https://openalex.org/W3148032049 |
| cited_by_count | 2 |
| counts_by_year[0].year | 2024 |
| counts_by_year[0].cited_by_count | 2 |
| locations_count | 1 |
| best_oa_location.id | doi:10.1609/aaai.v38i11.29173 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4210191458 |
| best_oa_location.source.issn | 2159-5399, 2374-3468 |
| best_oa_location.source.type | conference |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | 2159-5399 |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| best_oa_location.source.host_organization | https://openalex.org/P4310320058 |
| best_oa_location.source.host_organization_name | Association for the Advancement of Artificial Intelligence |
| best_oa_location.source.host_organization_lineage | https://openalex.org/P4310320058 |
| best_oa_location.source.host_organization_lineage_names | Association for the Advancement of Artificial Intelligence |
| best_oa_location.license | |
| best_oa_location.pdf_url | |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | journal-article |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | True |
| best_oa_location.is_published | True |
| best_oa_location.raw_source_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| best_oa_location.landing_page_url | http://doi.org/10.1609/aaai.v38i11.29173 |
| primary_location.id | doi:10.1609/aaai.v38i11.29173 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4210191458 |
| primary_location.source.issn | 2159-5399, 2374-3468 |
| primary_location.source.type | conference |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | 2159-5399 |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| primary_location.source.host_organization | https://openalex.org/P4310320058 |
| primary_location.source.host_organization_name | Association for the Advancement of Artificial Intelligence |
| primary_location.source.host_organization_lineage | https://openalex.org/P4310320058 |
| primary_location.source.host_organization_lineage_names | Association for the Advancement of Artificial Intelligence |
| primary_location.license | |
| primary_location.pdf_url | |
| primary_location.version | publishedVersion |
| primary_location.raw_type | journal-article |
| primary_location.license_id | |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Proceedings of the AAAI Conference on Artificial Intelligence |
| primary_location.landing_page_url | http://doi.org/10.1609/aaai.v38i11.29173 |
| publication_date | 2024-03-24 |
| publication_year | 2024 |
| referenced_works_count | 0 |
| abstract_inverted_index.9 | 126 |
| abstract_inverted_index.a | 4, 59, 88, 97, 117, 125 |
| abstract_inverted_index.In | 14 |
| abstract_inverted_index.We | 40, 142 |
| abstract_inverted_index.an | 42, 76 |
| abstract_inverted_index.as | 87 |
| abstract_inverted_index.by | 105, 169 |
| abstract_inverted_index.in | 9, 134 |
| abstract_inverted_index.is | 3, 20 |
| abstract_inverted_index.it | 107 |
| abstract_inverted_index.of | 23, 33, 61, 64, 85, 90, 121, 149 |
| abstract_inverted_index.on | 116, 146 |
| abstract_inverted_index.to | 68, 137, 159 |
| abstract_inverted_index.us | 67 |
| abstract_inverted_index.we | 54, 74, 95, 165 |
| abstract_inverted_index.15% | 131 |
| abstract_inverted_index.39M | 122 |
| abstract_inverted_index.and | 155 |
| abstract_inverted_index.for | 101, 111, 152 |
| abstract_inverted_index.our | 144, 157 |
| abstract_inverted_index.the | 31, 38, 82, 91, 103, 112 |
| abstract_inverted_index.two | 147 |
| abstract_inverted_index.yet | 6 |
| abstract_inverted_index.SOTA | 167 |
| abstract_inverted_index.Slow | 0 |
| abstract_inverted_index.beat | 166 |
| abstract_inverted_index.data | 19 |
| abstract_inverted_index.from | 37 |
| abstract_inverted_index.more | 21 |
| abstract_inverted_index.over | 49, 124 |
| abstract_inverted_index.rich | 70 |
| abstract_inverted_index.risk | 32 |
| abstract_inverted_index.runs | 30 |
| abstract_inverted_index.show | 129 |
| abstract_inverted_index.such | 15 |
| abstract_inverted_index.that | 80 |
| abstract_inverted_index.too, | 164 |
| abstract_inverted_index.upto | 130 |
| abstract_inverted_index.work | 158 |
| abstract_inverted_index.year | 127 |
| abstract_inverted_index.data, | 25 |
| abstract_inverted_index.drift | 2 |
| abstract_inverted_index.gains | 133, 145 |
| abstract_inverted_index.large | 50, 118, 170 |
| abstract_inverted_index.learn | 75 |
| abstract_inverted_index.model | 55, 79 |
| abstract_inverted_index.other | 138 |
| abstract_inverted_index.past. | 39 |
| abstract_inverted_index.using | 58 |
| abstract_inverted_index.which | 106 |
| abstract_inverted_index.First, | 53 |
| abstract_inverted_index.decay, | 65 |
| abstract_inverted_index.extend | 156 |
| abstract_inverted_index.future | 24 |
| abstract_inverted_index.losing | 34 |
| abstract_inverted_index.model. | 114 |
| abstract_inverted_index.nested | 98 |
| abstract_inverted_index.period | 128 |
| abstract_inverted_index.photos | 123 |
| abstract_inverted_index.recent | 18, 28 |
| abstract_inverted_index.robust | 139 |
| abstract_inverted_index.scorer | 78 |
| abstract_inverted_index.where, | 163 |
| abstract_inverted_index.Second, | 73 |
| abstract_inverted_index.capture | 69 |
| abstract_inverted_index.concept | 1 |
| abstract_inverted_index.dataset | 120 |
| abstract_inverted_index.forward | 109 |
| abstract_inverted_index.itself. | 93 |
| abstract_inverted_index.learned | 113 |
| abstract_inverted_index.machine | 11 |
| abstract_inverted_index.methods | 168 |
| abstract_inverted_index.mixture | 60, 84 |
| abstract_inverted_index.naively | 26 |
| abstract_inverted_index.problem | 8 |
| abstract_inverted_index.propose | 41, 96 |
| abstract_inverted_index.scorer, | 104 |
| abstract_inverted_index.towards | 45 |
| abstract_inverted_index.trends. | 72 |
| abstract_inverted_index.Finally, | 94 |
| abstract_inverted_index.accuracy | 135 |
| abstract_inverted_index.allowing | 66 |
| abstract_inverted_index.although | 17 |
| abstract_inverted_index.approach | 44 |
| abstract_inverted_index.compared | 136 |
| abstract_inverted_index.datasets | 151 |
| abstract_inverted_index.function | 89 |
| abstract_inverted_index.instance | 47, 56, 92 |
| abstract_inverted_index.learning | 12, 102, 140, 161 |
| abstract_inverted_index.margins. | 171 |
| abstract_inverted_index.multiple | 62 |
| abstract_inverted_index.recovers | 81 |
| abstract_inverted_index.relative | 132 |
| abstract_inverted_index.settings | 162 |
| abstract_inverted_index.systems. | 13 |
| abstract_inverted_index.temporal | 71 |
| abstract_inverted_index.training | 51 |
| abstract_inverted_index.transfer | 110 |
| abstract_inverted_index.valuable | 35 |
| abstract_inverted_index.windows. | 52 |
| abstract_inverted_index.auxiliary | 77 |
| abstract_inverted_index.balancing | 46 |
| abstract_inverted_index.continual | 160 |
| abstract_inverted_index.instances | 29 |
| abstract_inverted_index.learning, | 154 |
| abstract_inverted_index.maximizes | 108 |
| abstract_inverted_index.objective | 100 |
| abstract_inverted_index.practical | 10 |
| abstract_inverted_index.relevance | 57 |
| abstract_inverted_index.replicate | 143 |
| abstract_inverted_index.settings, | 16 |
| abstract_inverted_index.baselines. | 141 |
| abstract_inverted_index.importance | 48 |
| abstract_inverted_index.indicative | 22 |
| abstract_inverted_index.real-world | 119, 150 |
| abstract_inverted_index.timescales | 63, 86 |
| abstract_inverted_index.Experiments | 115 |
| abstract_inverted_index.appropriate | 83 |
| abstract_inverted_index.collections | 148 |
| abstract_inverted_index.information | 36 |
| abstract_inverted_index.ubiquitous, | 5 |
| abstract_inverted_index.optimization | 99 |
| abstract_inverted_index.prioritizing | 27 |
| abstract_inverted_index.under-studied | 7 |
| abstract_inverted_index.non-stationary | 153 |
| abstract_inverted_index.optimization-driven | 43 |
| cited_by_percentile_year.max | 96 |
| cited_by_percentile_year.min | 94 |
| countries_distinct_count | 0 |
| institutions_distinct_count | 2 |
| citation_normalized_percentile.value | 0.82831858 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | True |