Linear programming with online learning

We propose online decision strategies for time-dependent sequences of linear programs which use no distributional and minimal geometric assumptions about the data. These strategies are obtained through Vovk’s aggregating algorithm which combines recommendations from a given strategy pool. We establish an averageperformance bound for the resulting solution sequence.

• Type: Journal
• Year: 2007
• Where: ORL
• Authors: Tatsiana Levina, Yuri Levin, Jeff McGill, Mikhail Nediak
• Keywords: Minimal Geometric Assumptions, Online Decision Strategies, ORL 2007, Time-dependent Sequences