Latent Features for Algorithm Selection

The success and power of algorithm selection techniques has been empirically demonstrated on numerous occasions, most noticeably in the competition settings like those for SAT, CSP, MaxSAT, QBF, etc. Yet while there is now a plethora of competing approaches, all of them are dependent on the quality of a set of structural features they use to distinguish amongst the instances. Over the years, each domain has defined and refined its own set of features, yet at their core they are mostly a collection of everything that was considered useful in the past. As an alternative to this shotgun generation of features, this paper instead proposes a more systematic approach. Specifically, the paper shows how latent features gathered from matrix decomposition are enough for a linear model to achieve a level of performance comparable to a perfect Oracle portfolio. This information can, in turn, help guide researchers to the kinds of structural features they should be looking for, or even just identifying when such features are missing.

• Publication year

  2014

• Venue

  Symposium on Combinatorial Search

• Publication date

  2014-07-02

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1609/socs.v5i1.18324
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• MaxSAT by Improved Instance-Specific Algorithm Configuration

  2014cited by this paper
• Proteus: A Hierarchical Portfolio of Solvers and Transformations

  2013cited by this paper
• LLAMA: Leveraging Learning to Automatically Manage Algorithms

  2013cited by this paper
• A Latent Instrumental Variables Approach to Modeling Keyword Conversion in Paid Search Advertising

  2012cited by this paper
• Algorithm runtime prediction: Methods & evaluation

  2012cited by this paper
• An evaluation of machine learning in algorithm selection for search problems

  2012cited by this paper
• Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms

  2012cited by this paper
• Statistical identification of syndromes feature and structure of disease of western medicine based on general latent structure model

  2012cited by this paper
• Bayesian Nonparametric Latent Feature Models

  2011cited by this paper
• Feature Filtering for Instance-Specific Algorithm Configuration

  2011cited by this paper
• The Extraordinary SVD

  2011cited by this paper
• Algorithm Selection and Scheduling

  2011influential reference
• Fast Downward Stone Soup : A Baseline for Building Planner Portfolios

  2011cited by this paper
• SATzilla: Portfolio-based Algorithm Selection for SAT

  2008cited by this paper
• Using Case-based Reasoning in an Algorithm Portfolio for Constraint Solving ?

  2008influential reference
• Image Compression Using SVD

  2007cited by this paper
• Bayesian nonparametric latent feature models

  2007cited by this paper
• An algorithm for the selection problem

  1986cited by this paper
• Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence Algorithm Portfolios Based on Cost-Sensitive Hierarchical Clustering

  year unknowncited by this paper

• Data-driven closure model selection for multiphase CFD via matrix completion

  2025cites this paper
• Extreme Algorithm Selection With Dyadic Feature Representation

  2020cites this paper
• Algorithm Selection as Recommendation: From Collaborative Filtering to Dyad Ranking

  2018cites this paper
• Alors: An algorithm recommender system

  2017influential citation
• Probabilistic Matrix Factorization for Automated Machine Learning

  2017cites this paper
• Data sampling through collaborative filtering for algorithm selection

  2017cites this paper
• A biased random-key genetic algorithm for the container pre-marshalling problem

  2016cites this paper
• Search Portfolio with Sharing

  2016cites this paper
• Deep Learning for Algorithm Portfolios

  2016cites this paper
• Joint Model Choice and Hyperparameter Optimization with Factorized Multilayer Perceptrons

  2015cites this paper
• An Algorithm Selection Benchmark of the Container Pre-marshalling Problem

  2015cites this paper