Timezone: »
This poster has been moved from Monday #86 to Thursday #101.
Stochastic convex optimization is a basic and well studied primitive in machine learning. It is well known that convex and Lipschitz functions can be minimized efficiently using Stochastic Gradient Descent (SGD).The Normalized Gradient Descent (NGD) algorithm, is an adaptation of Gradient Descent, which updates according to the direction of the gradients, rather than the gradients themselves. In this paper we analyze a stochastic version of NGD and prove its convergence to a global minimum for a wider class of functions: we require the functions to be quasi-convex and locally-Lipschitz. Quasi-convexity broadens the concept of unimodality to multidimensions and allows for certain types of saddle points, which are a known hurdle for first-order optimization methods such as gradient descent. Locally-Lipschitz functions are only required to be Lipschitz in a small region around the optimum. This assumption circumvents gradient explosion, which is another known hurdle for gradient descent variants. Interestingly, unlike the vanilla SGD algorithm, the stochastic normalized gradient descent algorithm provably requires a minimal minibatch size.
Author Information
Elad Hazan (Princeton University)
Kfir Y. Levy (Technion)
Shai Shalev-Shwartz (Hebrew University)
More from the Same Authors
-
2022 Poster: Knowledge Distillation: Bad Models Can Be Good Role Models »
Gal Kaplun · Eran Malach · Preetum Nakkiran · Shai Shalev-Shwartz -
2021 : Q&A with Shai Shalev-Shwartz »
Shai Shalev-Shwartz -
2021 : Deep Learning: Success, Failure, and the Border between them, Shai Shalev-Shwartz »
Shai Shalev-Shwartz -
2021 Poster: Online Control of Unknown Time-Varying Dynamical Systems »
Edgar Minasyan · Paula Gradu · Max Simchowitz · Elad Hazan -
2021 Poster: Faster Neural Network Training with Approximate Tensor Operations »
Menachem Adelman · Kfir Levy · Ido Hakimi · Mark Silberstein -
2021 Poster: Multiclass Boosting and the Cost of Weak Learning »
Nataly Brukhim · Elad Hazan · Shay Moran · Indraneel Mukherjee · Robert Schapire -
2021 Poster: STORM+: Fully Adaptive SGD with Recursive Momentum for Nonconvex Optimization »
Kfir Levy · Ali Kavis · Volkan Cevher -
2020 Poster: Adaptive Sampling for Stochastic Risk-Averse Learning »
Sebastian Curi · Kfir Y. Levy · Stefanie Jegelka · Andreas Krause -
2020 Poster: Geometric Exploration for Online Control »
Orestis Plevrakis · Elad Hazan -
2020 Poster: Non-Stochastic Control with Bandit Feedback »
Paula Gradu · John Hallman · Elad Hazan -
2020 Poster: Online Agnostic Boosting via Regret Minimization »
Nataly Brukhim · Xinyi Chen · Elad Hazan · Shay Moran -
2019 : Logarithmic Regret for Online Control »
Naman Agarwal · Elad Hazan · Karan Singh -
2019 : Poster and Coffee Break 1 »
Aaron Sidford · Aditya Mahajan · Alejandro Ribeiro · Alex Lewandowski · Ali H Sayed · Ambuj Tewari · Angelika Steger · Anima Anandkumar · Asier Mujika · Hilbert J Kappen · Bolei Zhou · Byron Boots · Chelsea Finn · Chen-Yu Wei · Chi Jin · Ching-An Cheng · Christina Yu · Clement Gehring · Craig Boutilier · Dahua Lin · Daniel McNamee · Daniel Russo · David Brandfonbrener · Denny Zhou · Devesh Jha · Diego Romeres · Doina Precup · Dominik Thalmeier · Eduard Gorbunov · Elad Hazan · Elena Smirnova · Elvis Dohmatob · Emma Brunskill · Enrique Munoz de Cote · Ethan Waldie · Florian Meier · Florian Schaefer · Ge Liu · Gergely Neu · Haim Kaplan · Hao Sun · Hengshuai Yao · Jalaj Bhandari · James A Preiss · Jayakumar Subramanian · Jiajin Li · Jieping Ye · Jimmy Smith · Joan Bas Serrano · Joan Bruna · John Langford · Jonathan Lee · Jose A. Arjona-Medina · Kaiqing Zhang · Karan Singh · Yuping Luo · Zafarali Ahmed · Zaiwei Chen · Zhaoran Wang · Zhizhong Li · Zhuoran Yang · Ziping Xu · Ziyang Tang · Yi Mao · David Brandfonbrener · Shirli Di-Castro · Riashat Islam · Zuyue Fu · Abhishek Naik · Saurabh Kumar · Benjamin Petit · Angeliki Kamoutsi · Simone Totaro · Arvind Raghunathan · Rui Wu · Donghwan Lee · Dongsheng Ding · Alec Koppel · Hao Sun · Christian Tjandraatmadja · Mahdi Karami · Jincheng Mei · Chenjun Xiao · Junfeng Wen · Zichen Zhang · Ross Goroshin · Mohammad Pezeshki · Jiaqi Zhai · Philip Amortila · Shuo Huang · Mariya Vasileva · El houcine Bergou · Adel Ahmadyan · Haoran Sun · Sheng Zhang · Lukas Gruber · Yuanhao Wang · Tetiana Parshakova -
2019 Poster: Private Learning Implies Online Learning: An Efficient Reduction »
Alon Gonen · Elad Hazan · Shay Moran -
2019 Spotlight: Private Learning Implies Online Learning: An Efficient Reduction »
Alon Gonen · Elad Hazan · Shay Moran -
2019 Poster: A Domain Agnostic Measure for Monitoring and Evaluating GANs »
Paulina Grnarova · Kfir Y. Levy · Aurelien Lucchi · Nathanael Perraudin · Ian Goodfellow · Thomas Hofmann · Andreas Krause -
2019 Poster: UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization »
Ali Kavis · Kfir Y. Levy · Francis Bach · Volkan Cevher -
2019 Spotlight: UniXGrad: A Universal, Adaptive Algorithm with Optimal Guarantees for Constrained Optimization »
Ali Kavis · Kfir Y. Levy · Francis Bach · Volkan Cevher -
2019 Poster: Logarithmic Regret for Online Control »
Naman Agarwal · Elad Hazan · Karan Singh -
2019 Oral: Logarithmic Regret for Online Control »
Naman Agarwal · Elad Hazan · Karan Singh -
2018 Poster: Online Improper Learning with an Approximation Oracle »
Elad Hazan · Wei Hu · Yuanzhi Li · Zhiyuan Li -
2018 Poster: Online Adaptive Methods, Universality and Acceleration »
Kfir Y. Levy · Alp Yurtsever · Volkan Cevher -
2018 Poster: Online Learning of Quantum States »
Scott Aaronson · Xinyi Chen · Elad Hazan · Satyen Kale · Ashwin Nayak -
2018 Poster: Spectral Filtering for General Linear Dynamical Systems »
Elad Hazan · Holden Lee · Karan Singh · Cyril Zhang · Yi Zhang -
2018 Oral: Spectral Filtering for General Linear Dynamical Systems »
Elad Hazan · Holden Lee · Karan Singh · Cyril Zhang · Yi Zhang -
2017 Poster: Online to Offline Conversions, Universality and Adaptive Minibatch Sizes »
Kfir Levy -
2017 Poster: Linear Convergence of a Frank-Wolfe Type Algorithm over Trace-Norm Balls »
Zeyuan Allen-Zhu · Elad Hazan · Wei Hu · Yuanzhi Li -
2017 Poster: Learning Linear Dynamical Systems via Spectral Filtering »
Elad Hazan · Karan Singh · Cyril Zhang -
2017 Spotlight: Online Learning of Linear Dynamical Systems »
Elad Hazan · Karan Singh · Cyril Zhang -
2017 Spotlight: Linear Convergence of a Frank-Wolfe Type Algorithm over Trace-Norm Balls »
Zeyuan Allen-Zhu · Elad Hazan · Wei Hu · Yuanzhi Li -
2017 Poster: Non-monotone Continuous DR-submodular Maximization: Structure and Algorithms »
Yatao Bian · Kfir Levy · Andreas Krause · Joachim M Buhmann -
2016 Poster: Optimal Black-Box Reductions Between Optimization Objectives »
Zeyuan Allen-Zhu · Elad Hazan -
2016 Poster: k*-Nearest Neighbors: From Global to Local »
Oren Anava · Kfir Y. Levy -
2016 Poster: A Non-generative Framework and Convex Relaxations for Unsupervised Learning »
Elad Hazan · Tengyu Ma -
2016 Poster: The Limits of Learning with Missing Data »
Brian Bullins · Elad Hazan · Tomer Koren -
2015 Poster: Online Learning for Adversaries with Memory: Price of Past Mistakes »
Oren Anava · Elad Hazan · Shie Mannor -
2015 Poster: Fast Rates for Exp-concave Empirical Risk Minimization »
Tomer Koren · Kfir Y. Levy -
2015 Poster: Online Gradient Boosting »
Alina Beygelzimer · Elad Hazan · Satyen Kale · Haipeng Luo -
2014 Poster: Bandit Convex Optimization: Towards Tight Bounds »
Elad Hazan · Kfir Y. Levy -
2009 Poster: On Stochastic and Worst-case Models for Investing »
Elad Hazan · Satyen Kale -
2009 Oral: On Stochastic and Worst-case Models for Investing »
Elad Hazan · Satyen Kale -
2009 Poster: An Efficient Interior-Point Method for Minimum-Regret Learning in Online Convex Optimization »
Elad Hazan · Nimrod Megiddo -
2009 Spotlight: An Efficient Interior-Point Method for Minimum-Regret Learning in Online Convex Optimization »
Elad Hazan · Nimrod Megiddo -
2009 Poster: Beyond Convexity: Online Submodular Minimization »
Elad Hazan · Satyen Kale -
2007 Oral: Adaptive Online Gradient Descent »
Peter Bartlett · Elad Hazan · Sasha Rakhlin -
2007 Poster: Adaptive Online Gradient Descent »
Peter Bartlett · Elad Hazan · Sasha Rakhlin -
2007 Poster: Computational Equivalence of Fixed Points and No Regret Algorithms, and Convergence to Equilibria »
Elad Hazan · Satyen Kale