mlsanityData to 7 Oct 2026

Can We Do Better Than Random Start? The Power of Data Outsourcing

arXiv 2205.08098 ↗First posted 17 May 2022cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

Many organizations have access to abundant data but lack the computational power to process the data. While they can outsource the computational task to other facilities, there are various constraints on the amount of data that can be shared. It is natural to ask what can data outsourcing accomplish under such constraints. We address this question from a machine learning perspective. When training a model with optimization algorithms, the quality of the results often relies heavily on the points where the algorithms are initialized. Random start is one of the most popular methods to tackle this issue, but it can be computationally expensive and not feasible for organizations lacking computing resources. Based on three different scenarios, we propose simulation-based algorithms that can utilize a small amount of outsourced data to find good initial points accordingly. Under suitable regularity conditions, we provide theoretical guarantees showing the algorithms can find good initial points with high probability. We also conduct numerical experiments to demonstrate that our algorithms perform significantly better than the random start approach.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 22 of its 37 references that are in the map. First posted on arXiv 17 May 2022.

Builds on

22 of 37 references.

References · 22

  1. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,531
  2. 2019On the Convergence of FedAvg on Non-IID Datacited by 1,220
  3. 2016Optimization Methods for Large-Scale Machine Learningcited by 1,140
  4. 2017Non-convex learning via Stochastic Gradient Langevin Dynamics: a nonasymptotic analysiscited by 242
  5. 2015Non-asymptotic convergence analysis for the Unadjusted Langevin Algorithmcited by 195
  6. 2019Fixup Initialization: Residual Learning Without Normalizationcited by 194
  7. 2019SCAFFOLD: Stochastic Controlled Averaging for Federated Learningcited by 174
  8. 2013Minimizing Finite Sums with the Stochastic Average Gradientcited by 164
  9. 2020FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Datacited by 143
  10. 2018Log-concave sampling: Metropolis-Hastings algorithms are fastcited by 117
  11. 2018How to Start Training: The Effect of Initialization and Architecturecited by 109
  12. 2017Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimizationcited by 96
10 more
  1. 2016The Landscape of Empirical Risk for Non-convex Lossescited by 86
  2. 2013Accelerated Gradient Methods for Nonconvex Nonlinear and Stochastic Programmingcited by 74
  3. 2018Gradient Descent with Random Initialization: Fast Global Convergence for Nonconvex Phase Retrievalcited by 71
  4. 2018Sampling Can Be Faster Than Optimizationcited by 68
  5. 2017Natasha 2: Faster Non-Convex Optimization Than SGDcited by 49
  6. 2016Stochastic Quasi-Newton Methods for Nonconvex Stochastic Optimizationcited by 37
  7. 2017Beyond Log-concavity: Provable Guarantees for Sampling Multi-modal Distributions using Simulated Tempering Langevin Monte Carlocited by 33
  8. 2018Simulated Tempering Langevin Monte Carlo II: An Improved Proof using Soft Markov Chain Decompositioncited by 16
  9. 2019On Stationary-Point Hitting Time and Ergodicity of Stochastic Gradient Langevin Dynamicscited by 14
  10. 2020Replica Exchange for Non-Convex Optimizationcited by 10

Built on by

Not cited yet.