English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Decisions made by machine learning models can have lasting impacts, makinglong-term fairness a critical consideration. It has been observed that ignoringthe long-term effect and directly applying fairness criterion in staticsettings can actually worsen bias over time. To address biases in sequentialdecision-making, we introduce a long-term fairness concept named EqualLong-term Benefit Rate (ELBERT). This concept is seamlessly integrated into aMarkov Decision Process (MDP) to consider the future effects of actions onlong-term fairness, thus providing a unified framework for fair sequentialdecision-making problems. ELBERT effectively addresses the temporaldiscrimination issues found in previous long-term fairness notions.Additionally, we demonstrate that the policy gradient of Long-term Benefit Ratecan be analytically simplified to standard policy gradients. Thissimplification makes conventional policy optimization methods viable forreducing bias, leading to our bias mitigation approach ELBERT-PO. Extensiveexperiments across various diverse sequential decision-making environmentsconsistently reveal that ELBERT-PO significantly diminishes bias whilemaintaining high utility. Code is available athttps://github.com/umd-huang-lab/ELBERT.

Adapting Static Fairness to Sequential Decision-Making: Bias Mitigation Strategies towards Equal Long-term Benefit Rate