Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms
arXiv:2010.11443
Abstract
We study the problem of improving the performance of online algorithms by incorporating machine-learned predictions. The goal is to design algorithms that are both consistent and robust, meaning that the algorithm performs well when predictions are accurate and maintains worst-case guarantees. Such algorithms have been studied in a recent line of works due to Lykouris and Vassilvitskii (ICML '18) and Purohit et al (NeurIPS '18). They provide robustness-consistency trade-offs for a variety of online problems. However, they leave open the question of whether these trade-offs are tight, i.e., to what extent to such trade-offs are necessary. In this paper, we provide the first set of non-trivial lower bounds for competitive analysis using machine-learned predictions. We focus on the classic problems of ski-rental and non-clairvoyant scheduling and provide optimal trade-offs in various settings.
To appear at NeurIPS 2020
References in corpus (11)
- Learning Combinatorial Optimization Algorithms over Graphs
- Competitive Paging Algorithms
- Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
- Neural Combinatorial Optimization with Reinforcement Learning
- Attention, Learn to Solve Routing Problems!
- Improving Online Algorithms via ML Predictions
- Learning to Branch
- Learning Fast Algorithms for Linear Transforms Using Butterfly Factorizations
- Learning-Based Low-Rank Approximations
- (Learned) Frequency Estimation Algorithms under Zipfian Distribution
- Learning-Assisted Competitive Algorithms for Peak-Aware Energy Scheduling