TL;DR
This paper introduces a data-driven approach for designing online algorithms that balance worst-case guarantees with good average-case performance, demonstrated on knapsack and set cover problems.
Contribution
It develops a method to create online algorithms with strong worst-case guarantees that also learn from data to perform well on typical inputs.
Findings
Achieved competitive bounds for policy classes in online knapsack and set cover.
Demonstrated practical benefits through a case study on electric vehicle charging.
Balanced worst-case guarantees with data-driven performance improvements.
Abstract
The design of online algorithms has tended to focus on algorithms with worst-case guarantees, e.g., bounds on the competitive ratio. However, it is well-known that such algorithms are often overly pessimistic, performing sub-optimally on non-worst-case inputs. In this paper, we develop an approach for data-driven design of online algorithms that maintain near-optimal worst-case guarantees while also performing learning in order to perform well for typical inputs. Our approach is to identify policy classes that admit global worst-case guarantees, and then perform learning using historical data within the policy classes. We demonstrate the approach in the context of two classical problems, online knapsack and online set cover, proving competitive bounds for rich policy classes in each case. Additionally, we illustrate the practical implications via a case study on electric vehicle…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
