Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods
Abstract
In this paper, we consider a non-convex loss-minimization problem of learning Supervised PageRank models, which can account for features of nodes and edges. We propose gradient-based and random gradient-free methods to solve this problem. Our algorithms are based on the concept of an inexact oracle and unlike the state-of-the-art gradient-based method we manage to provide theoretically the convergence rate guarantees for both of them. Finally, we compare the performance of the proposed optimization methods with the state of the art applied to a ranking task.
Cite
Text
Bogolubsky et al. "Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods." Neural Information Processing Systems, 2016.Markdown
[Bogolubsky et al. "Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods." Neural Information Processing Systems, 2016.](https://mlanthology.org/neurips/2016/bogolubsky2016neurips-learning/)BibTeX
@inproceedings{bogolubsky2016neurips-learning,
title = {{Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods}},
author = {Bogolubsky, Lev and Dvurechenskii, Pavel and Gasnikov, Alexander and Gusev, Gleb and Nesterov, Yurii and Raigorodskii, Andrei M and Tikhonov, Aleksey and Zhukovskii, Maksim},
booktitle = {Neural Information Processing Systems},
year = {2016},
pages = {4914-4922},
url = {https://mlanthology.org/neurips/2016/bogolubsky2016neurips-learning/}
}