Abstract | ||
---|---|---|
In this paper we study inexact dumped Newton method implemented in a distributed environment. We start with an original DiSCO algorithm [Communication-Efficient Distributed Optimization of Self-Concordant Empirical Loss, Yuchen Zhang and Lin Xiao, 2015]. We will show that this algorithm may not scale well and propose an algorithmic modifications which will lead to less communications, better load-balancing and more efficient computation. We perform numerical experiments with an regularized empirical loss minimization instance described by a 273GB dataset. |
Year | Venue | Field |
---|---|---|
2016 | national conference on artificial intelligence | Mathematical optimization,Distributed Computing Environment,Computer science,Load balancing (computing),Loss minimization,Artificial intelligence,Data partitioning,Machine learning,Computation,Newton's method |
DocType | Volume | Citations |
Journal | abs/1603.05191 | 1 |
PageRank | References | Authors |
0.34 | 20 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chenxin Ma | 1 | 73 | 5.25 |
Martin Takác | 2 | 752 | 49.49 |