arXiv preprint ·
Distributed Grid Search for Hyperparameter Tuning
- Published
- Source
- arXiv preprint, posted before peer review
- Original
- Read it at arXiv
- Topics
Abstract
The performance of many distributed optimization algorithms over peer-to-peer networks strongly depends on the choice of their hyperparameters. However, existing selection rules often require global quantities, yield conservative choices, or are tied to specific algorithms and hyperparameters. We propose Distributed Grid Search (D-GS), a consensus-based search method that evaluates and selects hyperparameter configurations across the network, requiring only the ability to run the tuned algorithm and evaluate local gradients. We also propose a zoom-in variant, Distributed Zoom Grid Search (D-ZGS), which improves empirical performance at a reduced cost. We prove that both methods terminate in finite time with all agents holding the same hyperparameter configuration, and quantify the (sub)optimality of this configuration. Numerical experiments on linear and logistic regression show that D-GS closely matches a centralized reference and that tuning and deployment cost a fraction of deploying algorithms with hyperparameters tuned using theoretical results.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Optimization & Control
Local Budget Benefits, Population Commitment, and Solver Performance: A DGO Case Study with Controlled DE Transfer
Non-asymptotic analysis of SGLD with discontinuous stochastic gradients: a new route to portfolio CVaR minimization
Quantum Vehicle Routing Problem: a smart initialization strategy for ill-conditioned optimization landscapes
Conic Certification of Transformer Attention
Stochastic Mirror Descent under Dependent Observations
Get the few papers worth your time in Optimization & Control every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.