Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization

Recent studies have illustrated that stochastic gradient Markov Chain Monte Carlo techniques have a strong potential in non-convex optimization, where local and global convergence guarantees can be shown under certain conditions. By building up on this recent theory, in this study, we develop an asynchronous-parallel stochastic L-BFGS algorithm for non-convex optimization. The proposed algorithm is suitable for both distributed and shared-memory settings. We provide formal theoretical analysis and show that the proposed method achieves an ergodic convergence rate of O(1/√N) (N being the total number of iterations) and it can achieve a linear speedup under certain conditions. We perform several experiments on both synthetic and real datasets. The results support our theory and show that the proposed algorithm provides a significant speedup over the recently proposed synchronous distributed L-BFGS algorithm.

Citation

Simsekli , U , Yildiz , C , Nguyen , T H , Richard , G & Cemgil , A T 2018 , Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization . in Proceedings of the 35th International Conference on Machine Learning . Proceedings of Machine Learning Research , vol. 80 , JMLR , pp. 4681-4690 , International Conference on Machine Learning , Stockholm , Sweden , 10/07/2018 . < http://proceedings.mlr.press/v80/simsekli18a.html >

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-201907304448

Collections

[article-cris] Perustieteiden korkeakoulu / SCI

Show all metadata