Automatic construction of parallel portfolios via algorithm configuration

Abstract

Since 2004, increases in computational power described by Moore's law have substantially been realized in the form of additional cores rather than through faster clock speeds. To make effective use of modern hardware when solving hard computational problems, it is therefore necessary to employ parallel solution strategies. In this work, we demonstrate how effective parallel solvers for propositional satisfiability (SAT), one of the most widely studied NP-complete problems, can be produced automatically from any existing sequential, highly parametric SAT solver. Our Automatic Construction of Parallel Portfolios (ACPP) approach uses an automatic algorithm configuration procedure to identify a set of configurations that perform well when executed in parallel. Applied to two prominent SAT solvers, Lingeling and clasp, our ACPP procedure identified 8-core solvers that significantly outperformed their sequential counterparts on a diverse set of instances from the application and hard combinatorial category of the 2012 SAT Challenge. We further extended our ACPP approach to produce parallel portfolio solvers consisting of several different solvers by combining their configuration spaces. Applied to the component solvers of the 2012 SAT Challenge gold medal winning SAT Solver pfolioUZK, our ACPP procedures produced a significantly better-performing parallel SAT solver.

Generative Adversarial Construction of Parallel Portfolios

“…Hydra constructs a portfolio iteratively by finding a configuration in each iteration that maximizes marginal contribution to the current portfolio, while ISAC clusters the training instances based on features and independently runs an algorithm configurator on each cluster. The basic ideas of Hydra and ISAC were later adapted to be used in constructing parallel portfolios, thus resulting in two new approaches PARHYDRA and CLUSTERING [17]. Another key approach for constructing parallel portfolios is PCIT [18], which also adopts an instance grouping strategy such as CLUSTERING but will adjust the grouping by transferring instances between subsets in the construction process.…”

“…This is acceptable (and even appealing) since the available computing power has been rapidly becoming much cheaper than before. 1 Indeed, such approaches have been demonstrated to be both practical and effective in cases of constructing sequential solvers [12], [13]; sequential portfolios [14]- [16] (i.e., algorithm portfolios with selectors/scheduling); and parallel portfolios [17], [18].…”

Pitfalls and Best Practices in Algorithm Configuration

“…Nevertheless, configurators have already been successfully applied to such large configuration spaces: GGA++ has been used to optimize over 100 parameters of Lingeling (Ansótegui et al, 2015), irace has been used to optimize over 200 parameters of the mixed integer programming solver SCIP (López-Ibáñez & Stützle, 2014;Achterberg, 2009) and with SMAC , we have optimized configuration spaces with over 900 parameters (Lindauer, Hoos, Leyton-Brown, & Schaub, 2017a).…”

How we designed winning algorithms for abstract argumentation and which insight we attained

“…There is an extensive literature on portfolio approaches, and techniques to select online promising algorithms. While the exploration of portfolio-based approaches for dealing with argumentation problems is outside the scope of this paper, we refer the interested reader to [46,47] for an overview of the area, and to [48,49,50] for more general information about portfolio approaches in AI. On the other hand, it has to be remarked that the tractable classes mentioned above, while interesting from a computational complexity point of view, represent very specific instances from a knowledge representation perspective if one considers the structure of original arguments.…”