Investigating and Explaining the Frequency Bias in Image Classification Article Swipe
Zhiyu Lin
,
Yifei Gao
,
Jitao Sang
·
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.24963/ijcai.2022/101
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.24963/ijcai.2022/101
CNNs exhibit many behaviors different from humans, one of which is the capability of employing high-frequency components. This paper discusses the frequency bias phenomenon in image classification tasks: the high-frequency components are actually much less exploited than the low- and mid- frequency components. We first investigate the frequency bias phenomenon by presenting two observations on feature discrimination and learning priority. Furthermore, we hypothesize that (1) the spectral density, (2) class consistency directly affect the frequency bias. Specifically, our investigations verify that the spectral density of datasets mainly affects the learning priority, while the class consistency mainly affects the feature discrimination.
Related Topics
Concepts
Metadata
- Type
- article
- Language
- en
- Landing Page
- https://doi.org/10.24963/ijcai.2022/101
- https://www.ijcai.org/proceedings/2022/0101.pdf
- OA Status
- bronze
- Cited By
- 10
- References
- 14
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4285605803
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4285605803Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.24963/ijcai.2022/101Digital Object Identifier
- Title
-
Investigating and Explaining the Frequency Bias in Image ClassificationWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2022Year of publication
- Publication date
-
2022-07-01Full publication date if available
- Authors
-
Zhiyu Lin, Yifei Gao, Jitao SangList of authors in order
- Landing page
-
https://doi.org/10.24963/ijcai.2022/101Publisher landing page
- PDF URL
-
https://www.ijcai.org/proceedings/2022/0101.pdfDirect link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
bronzeOpen access status per OpenAlex
- OA URL
-
https://www.ijcai.org/proceedings/2022/0101.pdfDirect OA link when available
- Concepts
-
Consistency (knowledge bases), Feature (linguistics), Computer science, Pattern recognition (psychology), Artificial intelligence, Class (philosophy), Radio spectrum, Phenomenon, Physics, Telecommunications, Quantum mechanics, Philosophy, LinguisticsTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
10Total citation count in OpenAlex
- Citations by year (recent)
-
2025: 2, 2024: 5, 2023: 2, 2022: 1Per-year citation counts (last 5 years)
- References (count)
-
14Number of works referenced by this work
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4285605803 |
|---|---|
| doi | https://doi.org/10.24963/ijcai.2022/101 |
| ids.doi | https://doi.org/10.24963/ijcai.2022/101 |
| ids.openalex | https://openalex.org/W4285605803 |
| fwci | 1.1755504 |
| type | article |
| title | Investigating and Explaining the Frequency Bias in Image Classification |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | 723 |
| biblio.first_page | 717 |
| topics[0].id | https://openalex.org/T11689 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9994999766349792 |
| 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 | Adversarial Robustness in Machine Learning |
| topics[1].id | https://openalex.org/T11512 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9945999979972839 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1702 |
| topics[1].subfield.display_name | Artificial Intelligence |
| topics[1].display_name | Anomaly Detection Techniques and Applications |
| topics[2].id | https://openalex.org/T10775 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9937999844551086 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1707 |
| topics[2].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[2].display_name | Generative Adversarial Networks and Image Synthesis |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C2776436953 |
| concepts[0].level | 2 |
| concepts[0].score | 0.7446251511573792 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q5163215 |
| concepts[0].display_name | Consistency (knowledge bases) |
| concepts[1].id | https://openalex.org/C2776401178 |
| concepts[1].level | 2 |
| concepts[1].score | 0.6262835264205933 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q12050496 |
| concepts[1].display_name | Feature (linguistics) |
| concepts[2].id | https://openalex.org/C41008148 |
| concepts[2].level | 0 |
| concepts[2].score | 0.6234279274940491 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[2].display_name | Computer science |
| concepts[3].id | https://openalex.org/C153180895 |
| concepts[3].level | 2 |
| concepts[3].score | 0.5313893556594849 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q7148389 |
| concepts[3].display_name | Pattern recognition (psychology) |
| concepts[4].id | https://openalex.org/C154945302 |
| concepts[4].level | 1 |
| concepts[4].score | 0.5309975147247314 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[4].display_name | Artificial intelligence |
| concepts[5].id | https://openalex.org/C2777212361 |
| concepts[5].level | 2 |
| concepts[5].score | 0.5254520773887634 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q5127848 |
| concepts[5].display_name | Class (philosophy) |
| concepts[6].id | https://openalex.org/C92545706 |
| concepts[6].level | 2 |
| concepts[6].score | 0.43078917264938354 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q902174 |
| concepts[6].display_name | Radio spectrum |
| concepts[7].id | https://openalex.org/C50335755 |
| concepts[7].level | 2 |
| concepts[7].score | 0.4270133078098297 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q483247 |
| concepts[7].display_name | Phenomenon |
| concepts[8].id | https://openalex.org/C121332964 |
| concepts[8].level | 0 |
| concepts[8].score | 0.14049255847930908 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q413 |
| concepts[8].display_name | Physics |
| concepts[9].id | https://openalex.org/C76155785 |
| concepts[9].level | 1 |
| concepts[9].score | 0.10595542192459106 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q418 |
| concepts[9].display_name | Telecommunications |
| concepts[10].id | https://openalex.org/C62520636 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q944 |
| concepts[10].display_name | Quantum mechanics |
| concepts[11].id | https://openalex.org/C138885662 |
| concepts[11].level | 0 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q5891 |
| concepts[11].display_name | Philosophy |
| concepts[12].id | https://openalex.org/C41895202 |
| concepts[12].level | 1 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q8162 |
| concepts[12].display_name | Linguistics |
| keywords[0].id | https://openalex.org/keywords/consistency |
| keywords[0].score | 0.7446251511573792 |
| keywords[0].display_name | Consistency (knowledge bases) |
| keywords[1].id | https://openalex.org/keywords/feature |
| keywords[1].score | 0.6262835264205933 |
| keywords[1].display_name | Feature (linguistics) |
| keywords[2].id | https://openalex.org/keywords/computer-science |
| keywords[2].score | 0.6234279274940491 |
| keywords[2].display_name | Computer science |
| keywords[3].id | https://openalex.org/keywords/pattern-recognition |
| keywords[3].score | 0.5313893556594849 |
| keywords[3].display_name | Pattern recognition (psychology) |
| keywords[4].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[4].score | 0.5309975147247314 |
| keywords[4].display_name | Artificial intelligence |
| keywords[5].id | https://openalex.org/keywords/class |
| keywords[5].score | 0.5254520773887634 |
| keywords[5].display_name | Class (philosophy) |
| keywords[6].id | https://openalex.org/keywords/radio-spectrum |
| keywords[6].score | 0.43078917264938354 |
| keywords[6].display_name | Radio spectrum |
| keywords[7].id | https://openalex.org/keywords/phenomenon |
| keywords[7].score | 0.4270133078098297 |
| keywords[7].display_name | Phenomenon |
| keywords[8].id | https://openalex.org/keywords/physics |
| keywords[8].score | 0.14049255847930908 |
| keywords[8].display_name | Physics |
| keywords[9].id | https://openalex.org/keywords/telecommunications |
| keywords[9].score | 0.10595542192459106 |
| keywords[9].display_name | Telecommunications |
| language | en |
| locations[0].id | doi:10.24963/ijcai.2022/101 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4363608755 |
| locations[0].source.issn | |
| locations[0].source.type | conference |
| locations[0].source.is_oa | False |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence |
| locations[0].source.host_organization | |
| locations[0].source.host_organization_name | |
| locations[0].license | |
| locations[0].pdf_url | https://www.ijcai.org/proceedings/2022/0101.pdf |
| locations[0].version | publishedVersion |
| locations[0].raw_type | proceedings-article |
| locations[0].license_id | |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence |
| locations[0].landing_page_url | https://doi.org/10.24963/ijcai.2022/101 |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5069028297 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-8045-9556 |
| authorships[0].author.display_name | Zhiyu Lin |
| authorships[0].countries | CN |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I21193070 |
| authorships[0].affiliations[0].raw_affiliation_string | Beijing Jiaotong University, China |
| authorships[0].institutions[0].id | https://openalex.org/I21193070 |
| authorships[0].institutions[0].ror | https://ror.org/01yj56c84 |
| authorships[0].institutions[0].type | education |
| authorships[0].institutions[0].lineage | https://openalex.org/I21193070 |
| authorships[0].institutions[0].country_code | CN |
| authorships[0].institutions[0].display_name | Beijing Jiaotong University |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Zhiyu Lin |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | Beijing Jiaotong University, China |
| authorships[1].author.id | https://openalex.org/A5114804187 |
| authorships[1].author.orcid | https://orcid.org/0009-0006-0436-5930 |
| authorships[1].author.display_name | Yifei Gao |
| authorships[1].countries | CN |
| authorships[1].affiliations[0].institution_ids | https://openalex.org/I21193070 |
| authorships[1].affiliations[0].raw_affiliation_string | Beijing Jiaotong University, China |
| authorships[1].institutions[0].id | https://openalex.org/I21193070 |
| authorships[1].institutions[0].ror | https://ror.org/01yj56c84 |
| authorships[1].institutions[0].type | education |
| authorships[1].institutions[0].lineage | https://openalex.org/I21193070 |
| authorships[1].institutions[0].country_code | CN |
| authorships[1].institutions[0].display_name | Beijing Jiaotong University |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Yifei Gao |
| authorships[1].is_corresponding | False |
| authorships[1].raw_affiliation_strings | Beijing Jiaotong University, China |
| authorships[2].author.id | https://openalex.org/A5023834030 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-0699-3205 |
| authorships[2].author.display_name | Jitao Sang |
| authorships[2].countries | CN |
| authorships[2].affiliations[0].institution_ids | https://openalex.org/I21193070 |
| authorships[2].affiliations[0].raw_affiliation_string | Beijing Jiaotong University, China |
| authorships[2].affiliations[1].institution_ids | https://openalex.org/I4210136793 |
| authorships[2].affiliations[1].raw_affiliation_string | Peng Cheng Lab, Shenzhen 518066, China |
| authorships[2].institutions[0].id | https://openalex.org/I21193070 |
| authorships[2].institutions[0].ror | https://ror.org/01yj56c84 |
| authorships[2].institutions[0].type | education |
| authorships[2].institutions[0].lineage | https://openalex.org/I21193070 |
| authorships[2].institutions[0].country_code | CN |
| authorships[2].institutions[0].display_name | Beijing Jiaotong University |
| authorships[2].institutions[1].id | https://openalex.org/I4210136793 |
| authorships[2].institutions[1].ror | https://ror.org/03qdqbt06 |
| authorships[2].institutions[1].type | facility |
| authorships[2].institutions[1].lineage | https://openalex.org/I4210136793 |
| authorships[2].institutions[1].country_code | CN |
| authorships[2].institutions[1].display_name | Peng Cheng Laboratory |
| authorships[2].author_position | last |
| authorships[2].raw_author_name | Jitao Sang |
| authorships[2].is_corresponding | False |
| authorships[2].raw_affiliation_strings | Beijing Jiaotong University, China, Peng Cheng Lab, Shenzhen 518066, China |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://www.ijcai.org/proceedings/2022/0101.pdf |
| open_access.oa_status | bronze |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Investigating and Explaining the Frequency Bias in Image Classification |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T03:46:38.306776 |
| primary_topic.id | https://openalex.org/T11689 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9994999766349792 |
| 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 | Adversarial Robustness in Machine Learning |
| related_works | https://openalex.org/W4300450609, https://openalex.org/W4386931570, https://openalex.org/W2391010541, https://openalex.org/W2357367123, https://openalex.org/W4388930439, https://openalex.org/W2387276901, https://openalex.org/W2385953334, https://openalex.org/W2351303360, https://openalex.org/W3134118520, https://openalex.org/W2349223072 |
| cited_by_count | 10 |
| counts_by_year[0].year | 2025 |
| counts_by_year[0].cited_by_count | 2 |
| counts_by_year[1].year | 2024 |
| counts_by_year[1].cited_by_count | 5 |
| counts_by_year[2].year | 2023 |
| counts_by_year[2].cited_by_count | 2 |
| counts_by_year[3].year | 2022 |
| counts_by_year[3].cited_by_count | 1 |
| locations_count | 1 |
| best_oa_location.id | doi:10.24963/ijcai.2022/101 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4363608755 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | conference |
| best_oa_location.source.is_oa | False |
| 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 | Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence |
| best_oa_location.source.host_organization | |
| best_oa_location.source.host_organization_name | |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://www.ijcai.org/proceedings/2022/0101.pdf |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | proceedings-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 Thirty-First International Joint Conference on Artificial Intelligence |
| best_oa_location.landing_page_url | https://doi.org/10.24963/ijcai.2022/101 |
| primary_location.id | doi:10.24963/ijcai.2022/101 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4363608755 |
| primary_location.source.issn | |
| primary_location.source.type | conference |
| primary_location.source.is_oa | False |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence |
| primary_location.source.host_organization | |
| primary_location.source.host_organization_name | |
| primary_location.license | |
| primary_location.pdf_url | https://www.ijcai.org/proceedings/2022/0101.pdf |
| primary_location.version | publishedVersion |
| primary_location.raw_type | proceedings-article |
| primary_location.license_id | |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence |
| primary_location.landing_page_url | https://doi.org/10.24963/ijcai.2022/101 |
| publication_date | 2022-07-01 |
| publication_year | 2022 |
| referenced_works | https://openalex.org/W4308831279, https://openalex.org/W3034864980, https://openalex.org/W2194775991, https://openalex.org/W3118608800, https://openalex.org/W2809600522, https://openalex.org/W4226324275, https://openalex.org/W2774616426, https://openalex.org/W2108598243, https://openalex.org/W3039817009, https://openalex.org/W2992308087, https://openalex.org/W3034175346, https://openalex.org/W2765407302, https://openalex.org/W3176756041, https://openalex.org/W2985484909 |
| referenced_works_count | 14 |
| abstract_inverted_index.We | 43 |
| abstract_inverted_index.by | 50 |
| abstract_inverted_index.in | 24 |
| abstract_inverted_index.is | 10 |
| abstract_inverted_index.of | 8, 13, 84 |
| abstract_inverted_index.on | 54 |
| abstract_inverted_index.we | 61 |
| abstract_inverted_index.(1) | 64 |
| abstract_inverted_index.(2) | 68 |
| abstract_inverted_index.and | 39, 57 |
| abstract_inverted_index.are | 31 |
| abstract_inverted_index.one | 7 |
| abstract_inverted_index.our | 77 |
| abstract_inverted_index.the | 11, 20, 28, 37, 46, 65, 73, 81, 88, 92, 97 |
| abstract_inverted_index.two | 52 |
| abstract_inverted_index.CNNs | 0 |
| abstract_inverted_index.This | 17 |
| abstract_inverted_index.bias | 22, 48 |
| abstract_inverted_index.from | 5 |
| abstract_inverted_index.less | 34 |
| abstract_inverted_index.low- | 38 |
| abstract_inverted_index.many | 2 |
| abstract_inverted_index.mid- | 40 |
| abstract_inverted_index.much | 33 |
| abstract_inverted_index.than | 36 |
| abstract_inverted_index.that | 63, 80 |
| abstract_inverted_index.bias. | 75 |
| abstract_inverted_index.class | 69, 93 |
| abstract_inverted_index.first | 44 |
| abstract_inverted_index.image | 25 |
| abstract_inverted_index.paper | 18 |
| abstract_inverted_index.which | 9 |
| abstract_inverted_index.while | 91 |
| abstract_inverted_index.affect | 72 |
| abstract_inverted_index.mainly | 86, 95 |
| abstract_inverted_index.tasks: | 27 |
| abstract_inverted_index.verify | 79 |
| abstract_inverted_index.affects | 87, 96 |
| abstract_inverted_index.density | 83 |
| abstract_inverted_index.exhibit | 1 |
| abstract_inverted_index.feature | 55, 98 |
| abstract_inverted_index.humans, | 6 |
| abstract_inverted_index.actually | 32 |
| abstract_inverted_index.datasets | 85 |
| abstract_inverted_index.density, | 67 |
| abstract_inverted_index.directly | 71 |
| abstract_inverted_index.learning | 58, 89 |
| abstract_inverted_index.spectral | 66, 82 |
| abstract_inverted_index.behaviors | 3 |
| abstract_inverted_index.different | 4 |
| abstract_inverted_index.discusses | 19 |
| abstract_inverted_index.employing | 14 |
| abstract_inverted_index.exploited | 35 |
| abstract_inverted_index.frequency | 21, 41, 47, 74 |
| abstract_inverted_index.priority, | 90 |
| abstract_inverted_index.priority. | 59 |
| abstract_inverted_index.capability | 12 |
| abstract_inverted_index.components | 30 |
| abstract_inverted_index.phenomenon | 23, 49 |
| abstract_inverted_index.presenting | 51 |
| abstract_inverted_index.components. | 16, 42 |
| abstract_inverted_index.consistency | 70, 94 |
| abstract_inverted_index.hypothesize | 62 |
| abstract_inverted_index.investigate | 45 |
| abstract_inverted_index.Furthermore, | 60 |
| abstract_inverted_index.observations | 53 |
| abstract_inverted_index.Specifically, | 76 |
| abstract_inverted_index.classification | 26 |
| abstract_inverted_index.discrimination | 56 |
| abstract_inverted_index.high-frequency | 15, 29 |
| abstract_inverted_index.investigations | 78 |
| abstract_inverted_index.discrimination. | 99 |
| cited_by_percentile_year.max | 98 |
| cited_by_percentile_year.min | 89 |
| countries_distinct_count | 1 |
| institutions_distinct_count | 3 |
| sustainable_development_goals[0].id | https://metadata.un.org/sdg/10 |
| sustainable_development_goals[0].score | 0.5899999737739563 |
| sustainable_development_goals[0].display_name | Reduced inequalities |
| citation_normalized_percentile.value | 0.78823973 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | False |