Learning Supervised PageRank with Gradient-Based and Gradient-Free Optimization Methods

In this paper, we consider a non-convex loss-minimization problem of learning Supervised PageRank models, which can account for some properties not considered by classical approaches such as the classical PageRank model. 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 state-of-the-art gradient-based method we manage to provide theoretically the convergence rate guarantees for both of them. In particular, under the assumption of local convexity of the loss function, our random gradient-free algorithm guarantees decrease of the loss function value expectation. At the same time, we theoretically justify that without convexity assumption for the loss function our gradient-based algorithm allows to find a point where the stationary condition is fulfilled with a given accuracy. For both proposed optimization algorithms, we find the settings of hyperparameters which give the lowest complexity (i.e., the number of arithmetic operations needed to achieve the given accuracy of the solution of the loss-minimization problem). The resulting estimates of the complexity are also provided. Finally, we apply proposed optimization algorithms to the web page ranking problem and compare proposed and state-of-the-art algorithms in terms of the considered loss function.

[1]  A. Andrew Convergence of an Iterative Method for Derivatives of Eigensystems , 1978 .

[2]  A. Andrew Iterative Computation of Derivatives of Eigenvalues and Eigenvectors , 1979 .

[3]  M. KleinbergJon Authoritative sources in a hyperlinked environment , 1999 .

[4]  Taher H. Haveliwala Efficient Computation of PageRank , 1999 .

[5]  Rajeev Motwani,et al.  The PageRank Citation Ranking : Bringing Order to the Web , 1999, WWW 1999.

[6]  Jon Kleinberg,et al.  Authoritative sources in a hyperlinked environment , 1999, SODA '98.

[7]  Matthew Richardson,et al.  The Intelligent surfer: Probabilistic Combination of Link and Content Information in PageRank , 2001, NIPS.

[8]  Taher H. Haveliwala Topic-sensitive PageRank , 2002, IEEE Trans. Knowl. Data Eng..

[9]  Jennifer Widom,et al.  Scaling personalized web search , 2003, WWW '03.

[10]  Kevin S. McCurley,et al.  Ranking the web frontier , 2004, WWW '04.

[11]  Yurii Nesterov,et al.  Cubic regularization of Newton method and its global performance , 2006, Math. Program..

[12]  Tie-Yan Liu,et al.  BrowseRank: letting web users vote for page importance , 2008, SIGIR '08.

[13]  Lin Xiao,et al.  Optimal Algorithms for Online Convex Optimization with Multi-Point Bandit Feedback. , 2010, COLT 2010.

[14]  Brian D. Davison,et al.  Freshness matters: in flowers, food, and web authority , 2010, SIGIR.

[15]  Tie-Yan Liu,et al.  Semi-supervised ranking on very large graphs with rich metadata , 2011, KDD.

[16]  Jure Leskovec,et al.  Supervised random walks: predicting and recommending links in social networks , 2010, WSDM '11.

[17]  Yurii Nesterov,et al.  Efficiency of Coordinate Descent Methods on Huge-Scale Optimization Problems , 2012, SIAM J. Optim..

[18]  Saeed Ghadimi,et al.  Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic Programming , 2013, SIAM J. Optim..

[19]  Gleb Gusev,et al.  URL Redirection Accounting for Improving Link-Based Ranking Methods , 2013, ECIR.

[20]  Gleb Gusev,et al.  Fresh BrowseRank , 2013, SIGIR.

[21]  Yurii Nesterov,et al.  Gradient methods for minimizing composite functions , 2012, Mathematical Programming.

[22]  Alexander Gasnikov,et al.  Efficient randomized algorithms for PageRank problem , 2014, ArXiv.

[23]  Gleb Gusev,et al.  Supervised Nested PageRank , 2014, CIKM.

[24]  Yurii Nesterov,et al.  First-order methods of smooth convex optimization with inexact oracle , 2013, Mathematical Programming.

[25]  Yurii Nesterov,et al.  Finding the stationary states of Markov chains by iterative methods , 2015, Appl. Math. Comput..

[26]  Gorjan Alagic,et al.  #p , 2019, Quantum information & computation.

[27]  Yurii Nesterov,et al.  Random Gradient-Free Minimization of Convex Functions , 2015, Foundations of Computational Mathematics.