Falcon: Fair Active Learning using Multi-armed Bandits

Summary: Falcon: scalable fair active learning that boosts group fairness during dataset curation by using a postpone-on-mismatch trial-and-error sampler to target desired (protected,label) groups despite unknown labels. It encodes the informativeness vs postpone-rate trade-off as policies and uses adversarial multi-armed bandits to pick the best policy, yielding substantially better fairness–accuracy tradeoffs and efficiency. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 13956
Venue: VLDB
Year: 2024
Pagerank: 5.5013766e-05
Overall Rank: 8,043 | 44.82%
DOI: 10.14778/3641204.3641207

Incoming Non-self Citations Over Time

Authors

1. Ki Hyun Tae (Korea Advanced Institute of Science and Technology)
2. Hantian Zhang (Georgia Institute of Technology)
3. Jaeyoung Park (Korea Advanced Institute of Science and Technology)
4. Kexin Rong (Georgia Institute of Technology)
5. Steven Euijong Whang (Korea Advanced Institute of Science and Technology)

BibTeX Citation

@article{tae_vldb24,
        title = {{Falcon: Fair Active Learning using Multi-armed Bandits}},
        author = {Tae, Ki Hyun and Zhang, Hantian and Park, Jaeyoung and Rong, Kexin and Whang, Steven Euijong},
        journal = {PVLDB},
        series = {{VLDB} '24},
        volume = {17},
        number = {5},
        pages = {952--965},
        doi = {10.14778/3641204.3641207},
        url = {https://doi.org/10.14778/3641204.3641207},
        year = {2024}
}

Incoming Citations (Sorted by Pagerank)

Showing 3 of 3 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
9,601	Low Rank Learning for Offline Query Optimization	2025	SIGMOD	5.2487799e-05
9,773	Fair and Actionable Causal Prescription Ruleset	2025	SIGMOD	5.2209769e-05
10,239	Fair Data Pre-Processing with Imperfect Attribute Space	2026	SIGMOD	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 11 of 11 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
205	Snorkel: Rapid Training Data Creation with Weak Supervision	2018	VLDB	0.00025235185
410	SeeDB: Efficient Data-Driven Visualization Recommendations to Support Visual Analytics	2015	VLDB	0.0001890421
863	Interventional Fairness : Causal Database Repair for Algorithmic Fairness	2019	SIGMOD	0.00013531835
4,382	Slice Tuner: A Selective Data Acquisition Framework for Accurate and Fair Machine Learning Models	2021	SIGMOD	6.7331832e-05
4,658	OmniFair: A Declarative System for Model-Agnostic Group Fairness in Machine Learning	2021	SIGMOD	6.5817368e-05
4,808	Selective Data Acquisition in the Wild for Model Charging	2022	VLDB	6.4991553e-05
5,334	Operationalizing Individual Fairness with Pairwise Fair Representations	2020	VLDB	6.2622047e-05
5,789	Seiden: Revisiting Query Processing in Video Database Systems	2023	VLDB	6.0874367e-05
6,854	iFlipper: Label Flipping for Individual Fairness	2023	SIGMOD	5.753578e-05
7,490	mlwhatif: What If You Could Stop Re-Implementing Your Machine Learning Pipeline Analyses Over and Over?	2023	VLDB	5.6061425e-05
8,801	Effective Entity Augmentation By Querying External Data Sources	2023	VLDB	5.3687048e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	1,951	Interpretable Data-Based Explanations for Fairness Debugging	2022	SIGMOD
2	13,306	Interactive Fairness Auditing: Leveraging AVOIR for Dynamic Evaluation and Mitigation	2025	SIGMOD
3	13,458	DENOUNCER: Detection of Unfairness in Classifiers	2021	VLDB
4	10,239	Fair Data Pre-Processing with Imperfect Attribute Space	2026	SIGMOD
5	863	Interventional Fairness : Causal Database Repair for Algorithmic Fairness	2019	SIGMOD
6	6,709	Through the Data Management Lens: Experimental Analysis and Evaluation of Fair Classification	2022	SIGMOD
7	4,658	OmniFair: A Declarative System for Model-Agnostic Group Fairness in Machine Learning	2021	SIGMOD
8	6,421	Falcon: A Privacy-Preserving and Interpretable Vertical Federated Learning System	2023	VLDB
9	6,637	Causal Feature Selection for Algorithmic Fairness	2022	SIGMOD
10	3,905	Automated Feature Engineering for Algorithmic Fairness	2021	VLDB