Online Unrelated Machine Load Balancing with Predictions Revisited

Assumption 1 can be made with a loss of (1+ )-factor in the competitive ratio. If pi′,j pi,j ≥ m for some j ∈ J, i, i′ ∈Mj , we can change pi′,j to∞. If the job j is assigned to i′ in the optimum solution, we assign it to the machine i∗ with the minimum pi∗,j instead. Thus, the processing time of j is decreased by at least a factor of m . We apply the operation for all violations of the assumption. Then the makespan of a machine i will be increased by at most (m−1)T m/ ≤ T . This holds since the total processing time of machines other than i in the optimum solution is at most (m − 1)T . We also remark the procedure that guarantees the assumption can run online, as jobs are handled separately in the procedure.

[1]  Richard M. Karp,et al.  An optimal algorithm for on-line bipartite matching , 1990, STOC '90.

[2]  Klaus Jansen,et al.  On the Configuration-LP of the Restricted Assignment Problem , 2016, SODA.

[3]  Klaus Jansen,et al.  A Quasi-Polynomial Approximation for the Restricted Assignment Problem , 2017, IPCO.

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

[5]  Mohammad Mahdian,et al.  Online bipartite matching with random arrivals: an approach based on strongly factor-revealing LPs , 2011, STOC '11.

[6]  Amos Fiat,et al.  On-line routing of virtual circuits with applications to load balancing and machine scheduling , 1997, JACM.

[7]  Jan Karel Lenstra,et al.  Approximation algorithms for scheduling unrelated parallel machines , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[8]  Nikhil R. Devanur,et al.  Randomized Primal-Dual analysis of RANKING for Online BiPartite Matching , 2013, SODA.

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

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

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

[12]  Amin Saberi,et al.  Online stochastic matching: online actions based on offline statistics , 2010, SODA '11.

[13]  Debmalya Panigrahi,et al.  Customizing ML Predictions for Online Algorithms , 2020, ICML.

[14]  Joan Boyar,et al.  Online Algorithms with Advice: A Survey , 2016, SIGA.

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

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

[17]  Yossi Azar,et al.  The competitiveness of on-line assignments , 1992, SODA '92.

[18]  Morteza Zadimoghaddam,et al.  Proportional Allocation: Simple, Distributed, and Diverse Matching with High Entropy , 2018, ICML.

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

[20]  Ola Svensson,et al.  The Primal-Dual method for Learning Augmented Algorithms , 2020, NeurIPS.

[21]  Klaus Jansen,et al.  Approximation, Randomization and Combinatorial Optimization. , 2008 .

[22]  Sanjeev Khanna,et al.  On (1, ∊)-Restricted Assignment Makespan Minimization , 2014, SODA.

[23]  Nikhil R. Devanur Online algorithms with stochastic input , 2011, SECO.

[24]  Benjamin Moseley,et al.  Learnable and Instance-Robust Predictions for Online Matching, Flows and Load Balancing , 2020, ESA.

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

[26]  Ola Svensson,et al.  Santa Claus schedules jobs on unrelated machines , 2010, STOC '11.