A Bandit You Can Trust
Saved in:
| Title: | A Bandit You Can Trust |
|---|---|
| Language: | English |
| Authors: | Ethan Prihar, Adam Sales, Neil Heffernan |
| Source: | Grantee Submission. 2023 (ptation). |
| Peer Reviewed: | Y |
| Page Count: | 10 |
| Publication Date: | 2023 |
| Sponsoring Agency: | National Science Foundation (NSF) Institute of Education Sciences (ED) Department of Education (ED) Office of Elementary and Secondary Education (OESE) (ED), Education Innovation and Research (EIR) Office of Naval Research (ONR) (DOD) Federal Highway Administration (FHWA), National Highway Institute (NHI) |
| Contract Number: | 2118725 2118904 1950683 1917808 1931523 1940236 1917713 1903304 1822830 1759229 1724889 1636782 1535428 R305N210049 R305D210031 R305A170137 R305A170243 R305A180401 R305A120125 P200A180088 P200A150306 U411B190024 S411B210024 N000141812768 R44GM146483 |
| Document Type: | Speeches/Meeting Papers Reports - Research |
| Descriptors: | Trust (Psychology), Learning Management Systems, Learning Processes, Algorithms, Reinforcement, Individualized Instruction, Learning Analytics, Distance Education, Computer Assisted Instruction, Computer Software |
| DOI: | 10.1145/3565472.3592955 |
| Abstract: | This work proposes Dynamic Linear Epsilon-Greedy, a novel contextual multi-armed bandit algorithm that can adaptively assign personalized content to users while enabling unbiased statistical analysis. Traditional A/B testing and reinforcement learning approaches have trade-offs between empirical investigation and maximal impact on users. Our algorithm seeks to balance these objectives, allowing platforms to personalize content effectively while still gathering valuable data. Dynamic Linear Epsilon-Greedy was evaluated via simulation and an empirical study in the ASSISTments online learning platform. In simulation, Dynamic Linear Epsilon-Greedy performed comparably to existing algorithms and in ASSISTments, slightly increased students' learning compared to A/B testing. Data collected from its recommendations allowed for the identification of qualitative interactions, which showed high and low knowledge students benefited from different content. Dynamic Linear Epsilon-Greedy holds promise as a method to balance personalization with unbiased statistical analysis. All the data collected during the simulation and empirical study are publicly available at https://osf.io/zuwf7/. [This paper was published in: "UMAP '23: Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization," June 26-29, 2023.] |
| Abstractor: | As Provided |
| Notes: | https://osf.io/zuwf7 |
| IES Funded: | Yes |
| Entry Date: | 2023 |
| Accession Number: | ED636016 |
| Database: | ERIC |
| Abstract: | This work proposes Dynamic Linear Epsilon-Greedy, a novel contextual multi-armed bandit algorithm that can adaptively assign personalized content to users while enabling unbiased statistical analysis. Traditional A/B testing and reinforcement learning approaches have trade-offs between empirical investigation and maximal impact on users. Our algorithm seeks to balance these objectives, allowing platforms to personalize content effectively while still gathering valuable data. Dynamic Linear Epsilon-Greedy was evaluated via simulation and an empirical study in the ASSISTments online learning platform. In simulation, Dynamic Linear Epsilon-Greedy performed comparably to existing algorithms and in ASSISTments, slightly increased students' learning compared to A/B testing. Data collected from its recommendations allowed for the identification of qualitative interactions, which showed high and low knowledge students benefited from different content. Dynamic Linear Epsilon-Greedy holds promise as a method to balance personalization with unbiased statistical analysis. All the data collected during the simulation and empirical study are publicly available at https://osf.io/zuwf7/. [This paper was published in: "UMAP '23: Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization," June 26-29, 2023.] |
|---|---|
| DOI: | 10.1145/3565472.3592955 |
