Black-Box Reductions for Parameter-free Online Learning in Banach Spaces

We introduce several new black-box reductions that significantly improve the design of adaptive and parameter-free online learning algorithms by simplifying analysis, improving regret guarantees, and sometimes even improving runtime. We reduce parameter-free online learning to online exp-concave optimization, we reduce optimization in a Banach space to one-dimensional optimization, and we reduce optimization over a constrained domain to unconstrained optimization. All of our reductions run as fast as online gradient descent. We use our new techniques to improve upon the previously best regret bounds for parameter-free learning, and do so for arbitrary norms.

[1]  Nikhil R. Devanur,et al.  Online Auctions and Multi-scale Online Learning , 2017, EC.

[2]  Ashok Cutkosky,et al.  Online Convex Optimization with Unconstrained Domains and Losses , 2017, NIPS.

[3]  Francesco Orabona,et al.  Dimension-Free Exponentiated Gradient , 2013, NIPS.

[4]  Ambuj Tewari,et al.  On the Universality of Online Mirror Descent , 2011, NIPS.

[5]  Mehryar Mohri,et al.  Parameter-Free Online Learning via Model Selection , 2017, NIPS.

[6]  Karthik Sridharan,et al.  Adaptive Online Learning , 2015, NIPS.

[7]  Ambuj Tewari,et al.  Smoothness, Low Noise and Fast Rates , 2010, NIPS.

[8]  Roberto Lucchetti,et al.  Convexity and well-posed problems , 2006 .

[9]  Matthew J. Streeter,et al.  Less Regret via Online Conditioning , 2010, ArXiv.

[10]  Roi Livni,et al.  Affine-Invariant Online Optimization and the Low-rank Experts Problem , 2017, NIPS.

[11]  Francesco Orabona,et al.  Coin Betting and Parameter-Free Online Learning , 2016, NIPS.

[12]  Haipeng Luo,et al.  Efficient Second Order Online Learning by Sketching , 2016, NIPS.

[13]  Wojciech Kotlowski,et al.  Scale-Invariant Unconstrained Online Learning , 2017, ALT.

[14]  Yoav Freund,et al.  A Parameter-free Hedging Algorithm , 2009, NIPS.

[15]  Ashok Cutkosky,et al.  Online Learning Without Prior Information , 2017, COLT.

[16]  I. Pinelis ROSENTHAL-TYPE INEQUALITIES FOR MARTINGALES IN 2-SMOOTH BANACH SPACES ∗ , 2012, 1212.1912.

[17]  Karthik Sridharan,et al.  Online Learning: Sufficient Statistics and the Burkholder Method , 2018, COLT.

[18]  Claudio Gentile,et al.  A Second-Order Perceptron Algorithm , 2002, SIAM J. Comput..

[19]  Gábor Lugosi,et al.  Prediction, learning, and games , 2006 .

[20]  Shai Shalev-Shwartz,et al.  Online learning: theory, algorithms and applications (למידה מקוונת.) , 2007 .

[21]  Ashok Cutkosky,et al.  Stochastic and Adversarial Online Learning without Hyperparameters , 2017, NIPS.

[22]  Nikhil R. Devanur,et al.  Multi-scale Online Learning and its Applications to Online Auctions , 2017 .

[23]  Matthew J. Streeter,et al.  No-Regret Algorithms for Unconstrained Online Convex Optimization , 2012, NIPS.

[24]  Peter L. Bartlett,et al.  Adaptive Online Gradient Descent , 2007, NIPS.

[25]  Petr Hájek,et al.  Biorthogonal Systems in Banach Spaces , 2007 .

[26]  Francesco Orabona,et al.  Unconstrained Online Linear Learning in Hilbert Spaces: Minimax Algorithms and Normal Approximations , 2014, COLT.

[27]  H. Brendan McMahan,et al.  A survey of Algorithms and Analysis for Adaptive Online Learning , 2014, J. Mach. Learn. Res..

[28]  Martin Zinkevich,et al.  Online Convex Programming and Generalized Infinitesimal Gradient Ascent , 2003, ICML.

[29]  Wouter M. Koolen,et al.  MetaGrad: Multiple Learning Rates in Online Learning , 2016, NIPS.

[30]  Francesco Orabona,et al.  Simultaneous Model Selection and Optimization through Parameter-free Stochastic Learning , 2014, NIPS.

[31]  Elad Hazan,et al.  Logarithmic regret algorithms for online convex optimization , 2006, Machine Learning.

[32]  Tatiana Tommasi,et al.  Training Deep Networks without Learning Rates Through Coin Betting , 2017, NIPS.