Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms

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.

[1]  Christos Tzamos,et al.  Pandora's Box with Correlations: Learning and Approximation , 2020, 2020 IEEE 61st Annual Symposium on Foundations of Computer Science (FOCS).

[2]  Michael Mitzenmacher,et al.  Scheduling with Predictions and the Price of Misprediction , 2019, ITCS.

[3]  Joseph Naor,et al.  The Design of Competitive Online Algorithms via a Primal-Dual Approach , 2009, Found. Trends Theor. Comput. Sci..

[4]  Aditya Bhaskara,et al.  Online Learning with Imperfect Hints , 2020, ICML.

[5]  Morteza Zadimoghaddam,et al.  Simultaneous approximations for adversarial and stochastic online budgeted allocation , 2012, SODA.

[6]  Piotr Indyk,et al.  Learning-Based Low-Rank Approximations , 2019, NeurIPS.

[7]  Joan Bruna,et al.  Intriguing properties of neural networks , 2013, ICLR.

[8]  Google,et al.  Improving Online Algorithms via ML Predictions , 2024, NeurIPS.

[9]  Sergei Vassilvitskii,et al.  Revenue Optimization with Approximate Bid Predictions , 2017, NIPS.

[10]  Lawrence V. Snyder,et al.  Reinforcement Learning for Solving the Vehicle Routing Problem , 2018, NeurIPS.

[11]  Amin Saberi,et al.  Online Optimization with Uncertain Information , 2012, ACM Trans. Algorithms.

[12]  Silvio Lattanzi,et al.  Online Scheduling via Learned Weights , 2020, SODA.

[13]  Ravi Kumar,et al.  Improving Online Algorithms via ML Predictions , 2018, NeurIPS.

[14]  Vahab S. Mirrokni,et al.  Online Allocation with Traffic Spikes: Mixing Adversarial and Stochastic Models , 2015, EC.

[15]  Antonios Antoniadis,et al.  Secretary and Online Matching Problems with Machine Learned Advice , 2020, NeurIPS.

[16]  Christian Coester,et al.  Online Metric Algorithms with Untrusted Predictions , 2020, ICML.

[17]  Amos Fiat,et al.  Competitive Paging Algorithms , 1991, J. Algorithms.

[18]  Atri Rudra,et al.  Learning Fast Algorithms for Linear Transforms Using Butterfly Factorizations , 2019, ICML.

[19]  Samy Bengio,et al.  Neural Combinatorial Optimization with Reinforcement Learning , 2016, ICLR.

[20]  Piotr Indyk,et al.  (Learned) Frequency Estimation Algorithms under Zipfian Distribution , 2019, ArXiv.

[21]  Michael Mitzenmacher,et al.  The Supermarket Model with Known and Predicted Service Times , 2019, IEEE Transactions on Parallel and Distributed Systems.

[22]  Zhuwen Li,et al.  Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search , 2018, NeurIPS.

[23]  Ravi Kumar,et al.  Semi-Online Bipartite Matching , 2018, ITCS.

[24]  Le Song,et al.  2 Common Formulation for Greedy Algorithms on Graphs , 2018 .

[25]  David P. Woodruff,et al.  Learning-Augmented Data Stream Algorithms , 2020, ICLR.

[26]  Murali S. Kodialam,et al.  The constrained Ski-Rental problem and its application to online cloud cost optimization , 2013, 2013 Proceedings IEEE INFOCOM.

[27]  Piotr Indyk,et al.  Learning-Based Frequency Estimation Algorithms , 2018, ICLR.

[28]  Dhruv Rohatgi,et al.  Near-Optimal Bounds for Online Caching with Machine Learned Advice , 2019, SODA.

[29]  Markus Weimer,et al.  Learning To Solve Circuit-SAT: An Unsupervised Differentiable Approach , 2018, ICLR.

[30]  Jian Li,et al.  Learning-Assisted Competitive Algorithms for Peak-Aware Energy Scheduling , 2019, ArXiv.

[31]  David L. Dill,et al.  Learning a SAT Solver from Single-Bit Supervision , 2018, ICLR.

[32]  Sergei Vassilvitskii,et al.  Competitive caching with machine learned advice , 2018, ICML.

[33]  Max Welling,et al.  Attention, Learn to Solve Routing Problems! , 2018, ICLR.

[34]  Debmalya Panigrahi,et al.  Online Algorithms for Weighted Paging with Predictions , 2020, ICALP.

[35]  Adam Tauman Kalai,et al.  Learning to Prune: Speeding up Repeated Computations , 2019, COLT.

[36]  Anna R. Karlin,et al.  Competitive randomized algorithms for non-uniform problems , 1990, SODA '90.

[37]  Sreenivas Gollapudi,et al.  Online Algorithms for Rent-Or-Buy with Expert Advice , 2019, ICML.

[38]  Christoph Dürr,et al.  Online Computation with Untrusted Advice , 2019, ITCS.

[39]  Maria-Florina Balcan,et al.  Learning to Branch , 2018, ICML.

[40]  Alexander Wei,et al.  Better and Simpler Learning-Augmented Online Caching , 2020, APPROX-RANDOM.

[41]  Thomas P. Hayes,et al.  The adwords problem: online keyword matching with budgeted bidders under random permutations , 2009, EC '09.

[42]  Aranyak Mehta,et al.  Online Stochastic Matching: Beating 1-1/e , 2009, 2009 50th Annual IEEE Symposium on Foundations of Computer Science.

[43]  Edith Cohen,et al.  Composable Sketches for Functions of Frequencies: Beyond the Worst Case , 2020, ICML.

[44]  LEARNS OLD TRICKS,et al.  A new dog learns old tricks: RL finds classic optimization algorithms , 2018, ICLR.

[45]  Russell Bent,et al.  Online stochastic combinatorial optimization , 2006 .