Timezone: »
We develop methods for rapidly identifying important components of a convex optimization problem for the purpose of achieving fast convergence times. By considering a novel problem formulation—the minimization of a sum of piecewise functions—we describe a principled and general mechanism for exploiting piecewise linear structure in convex optimization. This result leads to a theoretically justified working set algorithm and a novel screening test, which generalize and improve upon many prior results on exploiting structure in convex optimization. In empirical comparisons, we study the scalability of our methods. We find that screening scales surprisingly poorly with the size of the problem, while our working set algorithm convincingly outperforms alternative approaches.
Author Information
Tyler Johnson (University of Washington)
Carlos Guestrin (Apple & University of Washington)
More from the Same Authors
-
2018 Poster: Learning to Optimize Tensor Programs »
Tianqi Chen · Lianmin Zheng · Eddie Yan · Ziheng Jiang · Thierry Moreau · Luis Ceze · Carlos Guestrin · Arvind Krishnamurthy -
2018 Spotlight: Learning to Optimize Tensor Programs »
Tianqi Chen · Lianmin Zheng · Eddie Yan · Ziheng Jiang · Thierry Moreau · Luis Ceze · Carlos Guestrin · Arvind Krishnamurthy -
2018 Poster: Training Deep Models Faster with Robust, Approximate Importance Sampling »
Tyler Johnson · Carlos Guestrin -
2016 : Invited talk, Carlos Guestrin »
Carlos Guestrin -
2014 Poster: Divide-and-Conquer Learning by Anchoring a Conical Hull »
Tianyi Zhou · Jeffrey A Bilmes · Carlos Guestrin -
2012 Demonstration: GraphLab: A Framework For Machine Learning in the Cloud »
Yucheng Low · Haijie Gu · Carlos Guestrin -
2011 Workshop: Big Learning: Algorithms, Systems, and Tools for Learning at Scale »
Joseph E Gonzalez · Sameer Singh · Graham Taylor · James Bergstra · Alice Zheng · Misha Bilenko · Yucheng Low · Yoshua Bengio · Michael Franklin · Carlos Guestrin · Andrew McCallum · Alexander Smola · Michael Jordan · Sugato Basu -
2011 Poster: Linear Submodular Bandits and their Application to Diversified Retrieval »
Yisong Yue · Carlos Guestrin -
2010 Poster: Evidence-Specific Structures for Rich Tractable CRFs »
Anton Chechetka · Carlos Guestrin -
2010 Poster: Inference with Multivariate Heavy-Tails in Linear Models »
Danny Bickson · Carlos Guestrin -
2009 Workshop: Learning with Orderings »
Tiberio Caetano · Carlos Guestrin · Jonathan Huang · Risi Kondor · Guy Lebanon · Marina Meila -
2009 Workshop: Large-Scale Machine Learning: Parallelism and Massive Datasets »
Alexander Gray · Arthur Gretton · Alexander Smola · Joseph E Gonzalez · Carlos Guestrin -
2009 Poster: Riffled Independence for Ranked Data »
Jonathan Huang · Carlos Guestrin -
2009 Spotlight: Riffled Independence for Ranked Data »
Jonathan Huang · Carlos Guestrin -
2007 Oral: Efficient Inference forDistributions on Permutations »
Jonathan Huang · Carlos Guestrin · Leonidas Guibas -
2007 Poster: Efficient Inference forDistributions on Permutations »
Jonathan Huang · Carlos Guestrin · Leonidas Guibas -
2007 Spotlight: Selecting Observations against Adversarial Objectives »
Andreas Krause · H. Brendan McMahan · Carlos Guestrin · Anupam Gupta -
2007 Poster: Selecting Observations against Adversarial Objectives »
Andreas Krause · H. Brendan McMahan · Carlos Guestrin · Anupam Gupta -
2007 Poster: Efficient Principled Learning of Thin Junction Trees »
Anton Chechetka · Carlos Guestrin -
2006 Poster: Distributed Inference in Dynamical Systems »
Stanislav Funiak · Carlos Guestrin · Mark A Paskin · Rahul Sukthankar