Comparing strings and assessing their similarity is a basic operation in many application domains of machine learning, such as in information retrieval, natural language processing and bioinformatics. The practitioner can choose from a large variety of available similarity measures for this task, each emphasizing different aspects of the string data. In this article, we present Harry, a small tool specifically designed for measuring the similarity of strings. Harry implements over 20 similarity measures, including common string distances and string kernels, such as the Levenshtein distance and the Subsequence kernel. The tool has been designed with efficiency in mind and allows for multi-threaded as well as distributed computing, enabling the analysis of large data sets of strings. Harry supports common data formats and thus can interface with analysis environments, such as Matlab, Pylab and Weka.
{"title":"Harry: A Tool for Measuring String Similarity","authors":"Konrad Rieck, Christian Wressnegger","doi":"10.5281/ZENODO.10074","DOIUrl":"https://doi.org/10.5281/ZENODO.10074","url":null,"abstract":"Comparing strings and assessing their similarity is a basic operation in many application domains of machine learning, such as in information retrieval, natural language processing and bioinformatics. The practitioner can choose from a large variety of available similarity measures for this task, each emphasizing different aspects of the string data. In this article, we present Harry, a small tool specifically designed for measuring the similarity of strings. Harry implements over 20 similarity measures, including common string distances and string kernels, such as the Levenshtein distance and the Subsequence kernel. The tool has been designed with efficiency in mind and allows for multi-threaded as well as distributed computing, enabling the analysis of large data sets of strings. Harry supports common data formats and thus can interface with analysis environments, such as Matlab, Pylab and Weka.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"37 1","pages":"9:1-9:5"},"PeriodicalIF":0.0,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79899781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ana M. Martínez, Geoffrey I. Webb, Shenglei Chen, Nayyar Zaidi
Ever increasing data quantity makes ever more urgent the need for highly scalable learners that have good classification performance. Therefore, an out-of-core learner with excellent time and space complexity, along with high expressivity (that is, capacity to learn very complex multivariate probability distributions) is extremely desirable. This paper presents such a learner. We propose an extension to the k-dependence Bayesian classifier (KDB) that discriminatively selects a sub-model of a full KDB classifier. It requires only one additional pass through the training data, making it a three-pass learner. Our extensive experimental evaluation on 16 large data sets reveals that this out-of-core algorithm achieves competitive classification performance, and substantially better training and classification time than state-of-the-art in-core learners such as random forest and linear and non-linear logistic regression.
{"title":"Scalable Learning of Bayesian Network Classifiers","authors":"Ana M. Martínez, Geoffrey I. Webb, Shenglei Chen, Nayyar Zaidi","doi":"10.5555/2946645.2946689","DOIUrl":"https://doi.org/10.5555/2946645.2946689","url":null,"abstract":"Ever increasing data quantity makes ever more urgent the need for highly scalable learners that have good classification performance. Therefore, an out-of-core learner with excellent time and space complexity, along with high expressivity (that is, capacity to learn very complex multivariate probability distributions) is extremely desirable. This paper presents such a learner. We propose an extension to the k-dependence Bayesian classifier (KDB) that discriminatively selects a sub-model of a full KDB classifier. It requires only one additional pass through the training data, making it a three-pass learner. Our extensive experimental evaluation on 16 large data sets reveals that this out-of-core algorithm achieves competitive classification performance, and substantially better training and classification time than state-of-the-art in-core learners such as random forest and linear and non-linear logistic regression.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"32 1","pages":"44:1-44:35"},"PeriodicalIF":0.0,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75291508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
José Miguel Hernández-Lobato, M. Gelbart, Ryan P. Adams, Matthew W. Hoffman, Zoubin Ghahramani
We present an information-theoretic framework for solving global black-box optimization problems that also have black-box constraints. Of particular interest to us is to efficiently solve problems with decoupled constraints, in which subsets of the objective and constraint functions may be evaluated independently. For example, when the objective is evaluated on a CPU and the constraints are evaluated independently on a GPU. These problems require an acquisition function that can be separated into the contributions of the individual function evaluations. We develop one such acquisition function and call it Predictive Entropy Search with Constraints (PESC). PESC is an approximation to the expected information gain criterion and it compares favorably to alternative approaches based on improvement in several synthetic and real-world problems. In addition to this, we consider problems with a mix of functions that are fast and slow to evaluate. These problems require balancing the amount of time spent in the meta-computation of PESC and in the actual evaluation of the target objective. We take a bounded rationality approach and develop a partial update for PESC which trades o_ accuracy against speed. We then propose a method for adaptively switching between the partial and full updates for PESC. This allows us to interpolate between versions of PESC that are efficient in terms of function evaluations and those that are efficient in terms of wall-clock time. Overall, we demonstrate that PESC is an effective algorithm that provides a promising direction towards a unified solution for constrained Bayesian optimization.
{"title":"A General Framework for Constrained Bayesian Optimization using Information-based Search","authors":"José Miguel Hernández-Lobato, M. Gelbart, Ryan P. Adams, Matthew W. Hoffman, Zoubin Ghahramani","doi":"10.17863/CAM.6477","DOIUrl":"https://doi.org/10.17863/CAM.6477","url":null,"abstract":"We present an information-theoretic framework for solving global black-box optimization problems that also have black-box constraints. Of particular interest to us is to efficiently solve problems with decoupled constraints, in which subsets of the objective and constraint functions may be evaluated independently. For example, when the objective is evaluated on a CPU and the constraints are evaluated independently on a GPU. These problems require an acquisition function that can be separated into the contributions of the individual function evaluations. We develop one such acquisition function and call it Predictive Entropy Search with Constraints (PESC). PESC is an approximation to the expected information gain criterion and it compares favorably to alternative approaches based on improvement in several synthetic and real-world problems. In addition to this, we consider problems with a mix of functions that are fast and slow to evaluate. These problems require balancing the amount of time spent in the meta-computation of PESC and in the actual evaluation of the target objective. We take a bounded rationality approach and develop a partial update for PESC which trades o_ accuracy against speed. We then propose a method for adaptively switching between the partial and full updates for PESC. This allows us to interpolate between versions of PESC that are efficient in terms of function evaluations and those that are efficient in terms of wall-clock time. Overall, we demonstrate that PESC is an effective algorithm that provides a promising direction towards a unified solution for constrained Bayesian optimization.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"6 1","pages":"160:1-160:53"},"PeriodicalIF":0.0,"publicationDate":"2015-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80620930","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Structured prediction is used in areas such as computer vision and natural language processing to predict structured outputs such as segmentations or parse trees. In these settings, prediction is performed by MAP inference or, equivalently, by solving an integer linear program. Because of the complex scoring functions required to obtain accurate predictions, both learning and inference typically require the use of approximate solvers. We propose a theoretical explanation to the striking observation that approximations based on linear programming (LP) relaxations are often tight on real-world instances. In particular, we show that learning with LP relaxed inference encourages integrality of training instances, and that tightness generalizes from train to test data.
{"title":"Train and Test Tightness of LP Relaxations in Structured Prediction","authors":"Ofer Meshi, M. Mahdavi, Adrian Weller, D. Sontag","doi":"10.17863/CAM.242","DOIUrl":"https://doi.org/10.17863/CAM.242","url":null,"abstract":"Structured prediction is used in areas such as computer vision and natural language processing to predict structured outputs such as segmentations or parse trees. In these settings, prediction is performed by MAP inference or, equivalently, by solving an integer linear program. Because of the complex scoring functions required to obtain accurate predictions, both learning and inference typically require the use of approximate solvers. We propose a theoretical explanation to the striking observation that approximations based on linear programming (LP) relaxations are often tight on real-world instances. In particular, we show that learning with LP relaxed inference encourages integrality of training instances, and that tightness generalizes from train to test data.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"1 1","pages":"13:1-13:34"},"PeriodicalIF":0.0,"publicationDate":"2015-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82473385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We propose some axioms for hierarchical clustering of probability measures and investigate their ramifications. The basic idea is to let the user stipulate the clusters for some elementary measures. This is done without the need of any notion of metric, similarity or dissimilarity. Our main results then show that for each suitable choice of user-defined clustering on elementary measures we obtain a unique notion of clustering on a large set of distributions satisfying a set of additivity and continuity axioms. We illustrate the developed theory by numerous examples including some with and some without a density.
{"title":"Towards an axiomatic approach to hierarchical clustering of measures","authors":"P. Thomann, Ingo Steinwart, Nico Schmid","doi":"10.5555/2789272.2886812","DOIUrl":"https://doi.org/10.5555/2789272.2886812","url":null,"abstract":"We propose some axioms for hierarchical clustering of probability measures and investigate their ramifications. The basic idea is to let the user stipulate the clusters for some elementary measures. This is done without the need of any notion of metric, similarity or dissimilarity. Our main results then show that for each suitable choice of user-defined clustering on elementary measures we obtain a unique notion of clustering on a large set of distributions satisfying a set of additivity and continuity axioms. We illustrate the developed theory by numerous examples including some with and some without a density.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"18 1","pages":"1949-2002"},"PeriodicalIF":0.0,"publicationDate":"2015-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81653314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Geramifard, Christoph Dann, Robert H. Klein, Will Dabney, J. How
RLPy is an object-oriented reinforcement learning software package with a focus on value-function-based methods using linear function approximation and discrete actions. The framework was designed for both educational and research purposes. It provides a rich library of fine-grained, easily exchangeable components for learning agents (e.g., policies or representations of value functions), facilitating recently increased specialization in reinforcement learning. RLPy is written in Python to allow fast prototyping, but is also suitable for large-scale experiments through its built-in support for optimized numerical libraries and parallelization. Code profiling, domain visualizations, and data analysis are integrated in a self-contained package available under the Modified BSD License at http://github.com/rlpy/rlpy. All of these properties allow users to compare various reinforcement learning algorithms with little effort.
{"title":"RLPy: a value-function-based reinforcement learning framework for education and research","authors":"A. Geramifard, Christoph Dann, Robert H. Klein, Will Dabney, J. How","doi":"10.5555/2789272.2886799","DOIUrl":"https://doi.org/10.5555/2789272.2886799","url":null,"abstract":"RLPy is an object-oriented reinforcement learning software package with a focus on value-function-based methods using linear function approximation and discrete actions. The framework was designed for both educational and research purposes. It provides a rich library of fine-grained, easily exchangeable components for learning agents (e.g., policies or representations of value functions), facilitating recently increased specialization in reinforcement learning. RLPy is written in Python to allow fast prototyping, but is also suitable for large-scale experiments through its built-in support for optimized numerical libraries and parallelization. Code profiling, domain visualizations, and data analysis are integrated in a self-contained package available under the Modified BSD License at http://github.com/rlpy/rlpy. All of these properties allow users to compare various reinforcement learning algorithms with little effort.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"78 1","pages":"1573-1578"},"PeriodicalIF":0.0,"publicationDate":"2015-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80057235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The density matrices are positively semi-definite Hermitian matrices of unit trace that describe the state of a quantum system. The goal of the paper is to develop minimax lower bounds on error rates of estimation of low rank density matrices in trace regression models used in quantum state tomography (in particular, in the case of Pauli measurements) with explicit dependence of the bounds on the rank and other complexity parameters. Such bounds are established for several statistically relevant distances, including quantum versions of Kullback-Leibler divergence (relative entropy distance) and of Hellinger distance (so called Bures distance), and Schatten $p$-norm distances. Sharp upper bounds and oracle inequalities for least squares estimator with von Neumann entropy penalization are obtained showing that minimax lower bounds are attained (up to logarithmic factors) for these distances.
{"title":"Optimal estimation of low rank density matrices","authors":"V. Koltchinskii, Dong Xia","doi":"10.5555/2789272.2886806","DOIUrl":"https://doi.org/10.5555/2789272.2886806","url":null,"abstract":"The density matrices are positively semi-definite Hermitian matrices of unit trace that describe the state of a quantum system. The goal of the paper is to develop minimax lower bounds on error rates of estimation of low rank density matrices in trace regression models used in quantum state tomography (in particular, in the case of Pauli measurements) with explicit dependence of the bounds on the rank and other complexity parameters. Such bounds are established for several statistically relevant distances, including quantum versions of Kullback-Leibler divergence (relative entropy distance) and of Hellinger distance (so called Bures distance), and Schatten $p$-norm distances. Sharp upper bounds and oracle inequalities for least squares estimator with von Neumann entropy penalization are obtained showing that minimax lower bounds are attained (up to logarithmic factors) for these distances.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"32 1","pages":"1757-1792"},"PeriodicalIF":0.0,"publicationDate":"2015-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82020503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We derive oracle inequalities for the problems of isotonic and convex regression using the combination of $Q$-aggregation procedure and sparsity pattern aggregation. This improves upon the previous results including the oracle inequalities for the constrained least squares estimator. One of the improvements is that our oracle inequalities are sharp, i.e., with leading constant 1. It allows us to obtain bounds for the minimax regret thus accounting for model misspecification, which was not possible based on the previous results. Another improvement is that we obtain oracle inequalities both with high probability and in expectation.
{"title":"Sharp oracle bounds for monotone and convex regression through aggregation","authors":"P. Bellec, A. Tsybakov","doi":"10.5555/2789272.2886809","DOIUrl":"https://doi.org/10.5555/2789272.2886809","url":null,"abstract":"We derive oracle inequalities for the problems of isotonic and convex regression using the combination of $Q$-aggregation procedure and sparsity pattern aggregation. This improves upon the previous results including the oracle inequalities for the constrained least squares estimator. One of the improvements is that our oracle inequalities are sharp, i.e., with leading constant 1. It allows us to obtain bounds for the minimax regret thus accounting for model misspecification, which was not possible based on the previous results. Another improvement is that we obtain oracle inequalities both with high probability and in expectation.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"49 1","pages":"1879-1892"},"PeriodicalIF":0.0,"publicationDate":"2015-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87563078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper introduces the Encog library for Java and C#, a scalable, adaptable, multiplatform machine learning framework that was 1st released in 2008. Encog allows a variety of machine learning models to be applied to datasets using regression, classification, and clustering. Various supported machine learning models can be used interchangeably with minimal recoding. Encog uses efficient multithreaded code to reduce training time by exploiting modern multicore processors. The current version of Encog can be downloaded from this http URL
{"title":"Encog: library of interchangeable machine learning models for Java and C#","authors":"Jeff Heaton","doi":"10.5555/2789272.2886789","DOIUrl":"https://doi.org/10.5555/2789272.2886789","url":null,"abstract":"This paper introduces the Encog library for Java and C#, a scalable, adaptable, multiplatform machine learning framework that was 1st released in 2008. Encog allows a variety of machine learning models to be applied to datasets using regression, classification, and clustering. Various supported machine learning models can be used interchangeably with minimal recoding. Encog uses efficient multithreaded code to reduce training time by exploiting modern multicore processors. The current version of Encog can be downloaded from this http URL","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"21 1","pages":"1243-1247"},"PeriodicalIF":0.0,"publicationDate":"2015-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86239936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. Addario-Berry, S. Bhamidi, Sébastien Bubeck, L. Devroye, G. Lugosi, R. Oliveira
In this paper we explore maximal deviations of large random structures from their typical behavior. We introduce a model for a high-dimensional random graph process and ask analogous questions to those of Vapnik and Chervonenkis for deviations of averages: how "rich" does the process have to be so that one sees atypical behavior. In particular, we study a natural process of ErdH{o}s-R'enyi random graphs indexed by unit vectors in $mathbb{R}^d$. We investigate the deviations of the process with respect to three fundamental properties: clique number, chromatic number, and connectivity. In all cases we establish upper and lower bounds for the minimal dimension $d$ that guarantees the existence of "exceptional directions" in which the random graph behaves atypically with respect to the property. For each of the three properties, four theorems are established, to describe upper and lower bounds for the threshold dimension in the subcritical and supercritical regimes.
{"title":"Exceptional rotations of random graphs: a VC theory","authors":"L. Addario-Berry, S. Bhamidi, Sébastien Bubeck, L. Devroye, G. Lugosi, R. Oliveira","doi":"10.5555/2789272.2886810","DOIUrl":"https://doi.org/10.5555/2789272.2886810","url":null,"abstract":"In this paper we explore maximal deviations of large random structures from their typical behavior. We introduce a model for a high-dimensional random graph process and ask analogous questions to those of Vapnik and Chervonenkis for deviations of averages: how \"rich\" does the process have to be so that one sees atypical behavior. In particular, we study a natural process of ErdH{o}s-R'enyi random graphs indexed by unit vectors in $mathbb{R}^d$. We investigate the deviations of the process with respect to three fundamental properties: clique number, chromatic number, and connectivity. In all cases we establish upper and lower bounds for the minimal dimension $d$ that guarantees the existence of \"exceptional directions\" in which the random graph behaves atypically with respect to the property. For each of the three properties, four theorems are established, to describe upper and lower bounds for the threshold dimension in the subcritical and supercritical regimes.","PeriodicalId":14794,"journal":{"name":"J. Mach. Learn. Res.","volume":"11 1","pages":"1893-1922"},"PeriodicalIF":0.0,"publicationDate":"2015-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80765627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：