Mark W. Stephenson, Saman Amarasinghe, Martín Martín et al. · 2003
Compiler writers have crafted many heuristics over the years to approximately solve NP-hard problems efficiently. Finding a heuristic that performs well on a broad range of applications is a tedious and difficult process. This paper introduces Meta Optimization, a methodology for automatically fine-tuning compiler heuristics. Meta Optimization uses machine-learning techniques to automatically search the space of compiler heuristics. Our techniques reduce compiler design complexity by relieving compiler writers of the tedium of heuristic tuning. Our machine-learning system uses an evolutionary algorithm to automatically find effective compiler heuristics. We present promising experimental results. In one mode of operation Meta Optimization creates application-specific heuristics which often result in impressive speedups. For hyperblock formation, one optimization we present in this paper, we obtain an average speedup of 23% (up to 73%) for the applications in our suite. Furthermore, by evolving a compiler's heuristic over several benchmarks, we can create effective, general-purpose heuristics. The best general-purpose heuristic our system found for hyperblock formation improved performance by an average of 25% on our training set, and 9% on a completely unrelated test set. We demonstrate the efficacy of our techniques on three different optimizations in this paper: hyperblock formation, register allocation, and data prefetching.
Cited by 6
Oleg Kovářík, Richard Málek · 2013
Abstract. Meta-learning is a method of improving results of algorithm by learning from metafeatures which describe problem instances and from results produced by various algorithms on these instances. In this project we tried to apply this idea, which was already proved to be useful in machine learning, to combinatorial optimization. We have developed a general software tool called SEAGE to extract meta-features and tune parameters of heuristics for several combinatorial problems. We have also developed a specialized tool Pheromone for extraction of meta-features for Traveling Salesman Problem (TSP) [1], which serves as a benchmark for combinatorial optimization. Source codes and database with extracted metafeatures and results of optimization are available online as open source. We show preliminary results of meta-learning on results for several algorithms and large set of TSP instances. These results indicate that meta-optimization can improve quality of solution by automatic algorithm recommendation.
Cited by 0
Daniel Huang, Junwei Lucas Bao, Jean-Baptiste Tristan · The Journal of Chemical Physics · 2022
Recent work has demonstrated the promise of using machine-learned surrogates, in particular, Gaussian process (GP) surrogates, in reducing the number of electronic structure calculations (ESCs) needed to perform surrogate model based (SMB) geometry optimization. In this paper, we study geometry meta-optimization with GP surrogates where a SMB optimizer additionally learns from its past "experience" performing geometry optimization. To validate this idea, we start with the simplest setting where a geometry meta-optimizer learns from previous optimizations of the same molecule with different initial-guess geometries. We give empirical evidence that geometry meta-optimization with GP surrogates is effective and requires less tuning compared to SMB optimization with GP surrogates on the ANI-1 dataset of off-equilibrium initial structures of small organic molecules. Unlike SMB optimization where a surrogate should be immediately useful for optimizing a given geometry, a surrogate in geometry meta-optimization has more flexibility because it can distribute its ESC savings across a set of geometries. Indeed, we find that GP surrogates that preserve rotational invariance provide increased marginal ESC savings across geometries. As a more stringent test, we also apply geometry meta-optimization to conformational search on a hand-constructed dataset of hydrocarbons and alcohols. We observe that while SMB optimization and geometry meta-optimization do save on ESCs, they also tend to miss higher energy conformers compared to standard geometry optimization. We believe that further research into characterizing the divergence between GP surrogates and potential energy surfaces is critical not only for advancing geometry meta-optimization but also for exploring the potential of machine-learned surrogates in geometry optimization in general.
Cited by 5
Lang, Andreas, Stanley, Kenneth · OSF Preprints (OSF Preprints) · 2016
This paper introduces a meta-optimization algorithm called NeuroEvolutionary Meta-Optimization (NEMO) that evolves an algorithm targeted at optimizing only within a specific problem class. More specifically, a form of neural network is evolved that acts as the controller of a kind of optimization algorithm that can potentially exploit problem class-specific structure. NEMO is demonstrated on several benchmark problems that confirm its ability to succeed on problems within the class on which it is trained. The key implication is that it is indeed possible to evolve this kind of meta-optimizer with a neural network-like structure, opening up a promising research direction in automatically evolving such class-specific optimizers.
Cited by 0Open access
Kenneth O. Stanley, Andreas Lang · OSF Preprints (OSF Preprints) · 2016
This paper introduces a meta-optimization algorithm called NeuroEvolutionary Meta-Optimization (NEMO) that evolves an algorithm targeted at optimizing only within a specific problem class. More specifically, a form of neural network is evolved that acts as the controller of a kind of optimization algorithm that can potentially exploit problem class-specific structure. NEMO is demonstrated on several benchmark problems that confirm its ability to succeed on problems within the class on which it is trained. The key implication is that it is indeed possible to evolve this kind of meta-optimizer with a neural network-like structure, opening up a promising research direction in automatically evolving such class-specific optimizers.
Cited by 0
Gottvald, Aleš · Digital Repository (National Repository of Grey Literature) · 1995
Cited by 1
Miquel Noguer I Alonso · SSRN Electronic Journal · 2025
Cited by 0Open access
Adnan Ahmed Khan, Sajid Bashir, Muhammad Naeem et al. · 2010
Cited by 0
2016
Cited by 0
Карпенко, Svianadze · Science and Education of the Bauman MSTU · 2015
Cited by 12Open access
С. В. Грошев, Anatoly Karpenko · On-line Journal Naukovedenie · 2016
Cited by 1
Manuel Laguna College · 2007
This article describes the origin and significant developments associated with the field of metaheuristics as they relate to global optimization. Meta-heuristics provide a means for approximately solving complex optimization problems. These methods are designed to search for global optima, however, they cannot guarantee that the best solution found after termination criteria are satisfied is indeed and global optimal solution to the problem. Experimental testing of meta-heuristic implementations show that the search strategies embedded in such procedures are capable of finding solutions of high quality to hard problems in industry, business and science
Cited by 0
Enzhi Zhang, Bochen Dong, Mohamed Wahib et al. · 2023
This paper proposes a method called Meta Gener-ative Data Augmentation Optimization (MGDAO) to overcome limitations in existing data augmentation techniques. While traditional data augmentation methods have relied on expert intuition to determine effective transformations, recent approaches have attempted to generate data augmentation strategies automatically. However, these automatic methods can still suffer from limited operation sets, high computational costs, or difficulty in training conditional generative models. To address these issues, MGDAO replaces the limited operations space in the AutoAugment series with a deep-style generator and replaces the discriminator in a generative adversarial model with the validation loss of the target model. The generator learns to transform the data from the training domain to the validation data domain. It is further used to generate data-augmented samples to train the target model and reduce the validation loss. Experiments on few-shot image classification benchmarks show that MGDAO achieves competitive results compared to existing data auto-augmentation methods.
Cited by 0
Xinyi Chen, Elad Hazan · 2023
Cited by 0
Prajna Kunche, K. V. V. S. Reddy · Springer briefs in electrical and computer engineering · 2016
Cited by 6
Kyle Erwin, Andries P. Engelbrecht · Soft Computing · 2023
Abstract Portfolio optimization has been studied extensively by researchers in computer science and finance, with new and novel work frequently published. Traditional methods, such as quadratic programming, are not computationally effective for solving complex portfolio models. For example, portfolio models with constraints that introduce nonlinearity and non-convexity (such as boundary constraints and cardinality constraints) are NP-Hard. As a result, researchers often use meta-heuristic approaches to approximate optimal solutions in an efficient manner. This paper conducts a comprehensive review of over 140 papers that have applied evolutionary and swarm intelligence algorithms to the portfolio optimization problem. These papers are categorized by the type of portfolio optimization problem considered, i.e., unconstrained or constrained, and are further categorized by single-objective and multi-objective approaches. Furthermore, the various portfolio models used, as well as the constraints, objectives, and properties in which they differ, are also discussed in a detailed analysis. Based on the findings of the reviewed work, guidance for future research in portfolio optimization is given. Possible areas for future work include dynamic portfolio optimization, predictive pricing, the further investigation of multi-objective approaches.
Cited by 45Open access
Al-Shaibani, Maged S. · Code Ocean · 2023
This capsule contains the code for the hybrid meta-heuristic optimization algorithm
Cited by 0Open access
Aggressive Tail Splitting, Michael Rock, Andreas Koch · 2004
Several optimization techniques are hindered by uncertainties about the control flow in a program, which can generally not be determined by static methods at compile time. We present a novel approach that aims to alleviate this limitation by explicitly enumerating all control paths through a loop nest, thus allowing wider and more efficient use of standard optimization passes later. Our analysis will address the possible explosion in code size, which despite high theoretical upper bounds appears to be acceptable in practice, as well as performance gains both for RISC and CISC target processors.
Cited by 0
Liqin Song · 2011
I would like to thank my supervisor, Dr
Cited by 0Open access
Igor Kuszczak, Miguel A. Bessa, Gawel Kus et al. · Zenodo (CERN European Organization for Nuclear Research) · 2025
This is a dataset accompanying the publication "Meta-neural Topology Optimization: Knowledge Infusion with Meta-learning".
Cited by 0Open access