-

 

 

 




Optimization Online





 

Bolstering Stochastic Gradient Descent with Model Building

S. Ilker Birbil(s.i.birbil***at***uva.nl)
Ozgur Martin(ozgur.martin***at***msgsu.edu.tr)
Gonenc Onay(onayg***at***posteo.net)
Figen Oztoprak(figen.topkaya***at***artelys.com)

Abstract: Stochastic gradient descent method and its variants constitute the core optimization algorithms that achieve good convergence rates for solving machine learning problems. These rates are obtained especially when these algorithms are fine-tuned for the application at hand. Although this tuning process can require large computational costs, recent work has shown that these costs can be reduced by line search methods that iteratively adjust the stepsize. We propose an alternative approach to stochastic line search by using a new algorithm based on forward step model building. This model building step incorporates a second-order information that allows adjusting not only the stepsize but also the search direction. Noting that deep learning model parameters come in groups (layers of tensors), our method builds its model and calculates a new step for each parameter group. This novel diagonalization approach makes the selected step lengths adaptive. We provide convergence rate analysis, and experimentally show that the proposed algorithm achieves faster convergence and better generalization in most problems. Moreover, our experiments show that the proposed method is quite robust as it converges for a wide range of initial stepsizes.

Keywords: model building; second-order information; stochastic gradient descent; convergence analysis

Category 1: Nonlinear Optimization (Unconstrained Optimization )

Citation:

Download: [PDF]

Entry Submitted: 11/13/2021
Entry Accepted: 11/13/2021
Entry Last Modified: 11/13/2021

Modify/Update this entry


  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository

 

Submit
Update
Policies
Coordinator's Board
Classification Scheme
Credits
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society