The Experts below are selected from a list of 19527 Experts worldwide ranked by ideXlab platform
Vladimir Vapnik - One of the best experts on this subject based on the ideXlab platform.
-
vicinal Risk Minimization
Neural Information Processing Systems, 2000Co-Authors: Olivier Chapelle, Leon Bottou, Jason Weston, Vladimir VapnikAbstract:The Vicinal Risk Minimization principle establishes a bridge between generative models and methods derived from the Structural Risk Minimization Principle such as Support Vector Machines or Statistical Regularization. We explain how VRM provides a framework which integrates a number of existing algorithms, such as Parzen windows, Support Vector Machines, Ridge Regression, Constrained Logistic Classifiers and Tangent-Prop. We then show how the approach implies new algorithms for solving problems usually associated with generative models. New algorithms are described for dealing with pattern recognition problems with very different pattern distributions and dealing with unlabeled data. Preliminary empirical results are presented.
-
NIPS - Vicinal Risk Minimization
2000Co-Authors: Olivier Chapelle, Leon Bottou, Jason Weston, Vladimir VapnikAbstract:The Vicinal Risk Minimization principle establishes a bridge between generative models and methods derived from the Structural Risk Minimization Principle such as Support Vector Machines or Statistical Regularization. We explain how VRM provides a framework which integrates a number of existing algorithms, such as Parzen windows, Support Vector Machines, Ridge Regression, Constrained Logistic Classifiers and Tangent-Prop. We then show how the approach implies new algorithms for solving problems usually associated with generative models. New algorithms are described for dealing with pattern recognition problems with very different pattern distributions and dealing with unlabeled data. Preliminary empirical results are presented.
-
The Vicinal Risk Minimization Principle and the SVMs
The Nature of Statistical Learning Theory, 2000Co-Authors: Vladimir VapnikAbstract:In this chapter we introduce a new principle for minimizing the expected Risk called the vicinal Risk Minimization (VRM) principle.’ We use this principle for solving our main problems: pattern recognition, regression estimation, and density estimation.
-
structural Risk Minimization for character recognition
Neural Information Processing Systems, 1991Co-Authors: Isabelle Guyon, Vladimir Vapnik, Bernhard E Boser, Leon Bottou, Sara A SollaAbstract:The method of Structural Risk Minimization refers to tuning the capacity of the classifier to the available amount of training data. This capacity is influenced by several factors, including: (1) properties of the input space, (2) nature and structure of the classifier, and (3) learning algorithm. Actions based on these three factors are combined here to control the capacity of linear classifiers and improve generalization on the problem of handwritten digit recognition.
-
NIPS - Structural Risk Minimization for Character Recognition
1991Co-Authors: Isabelle Guyon, Vladimir Vapnik, Bernhard E Boser, Leon Bottou, Sara A SollaAbstract:The method of Structural Risk Minimization refers to tuning the capacity of the classifier to the available amount of training data. This capacity is influenced by several factors, including: (1) properties of the input space, (2) nature and structure of the classifier, and (3) learning algorithm. Actions based on these three factors are combined here to control the capacity of linear classifiers and improve generalization on the problem of handwritten digit recognition.
Peter Orbanz - One of the best experts on this subject based on the ideXlab platform.
-
empirical Risk Minimization and stochastic gradient descent for relational data
International Conference on Artificial Intelligence and Statistics, 2019Co-Authors: Victor Veitch, Morgane Austern, Wenda Zhou, David M Blei, Peter OrbanzAbstract:Empirical Risk Minimization is the main tool for prediction problems, but its extension to relational data remains unsolved. We solve this problem using recent ideas from graph sampling theory to (i) define an empirical Risk for relational data and (ii) obtain stochastic gradients for this empirical Risk that are automatically unbiased. This is achieved by considering the method by which data is sampled from a graph as an explicit component of model design. By integrating fast implementations of graph sampling schemes with standard automatic differentiation tools, we provide an efficient turnkey solver for the Risk Minimization problem. We establish basic theoretical properties of the procedure. Finally, we demonstrate relational ERM with application to two non-standard problems: one-stage training for semi-supervised node classification, and learning embedding vectors for vertex attributes. Experiments confirm that the turnkey inference procedure is effective in practice, and that the sampling scheme used for model specification has a strong effect on model performance.
-
empirical Risk Minimization and stochastic gradient descent for relational data
arXiv: Machine Learning, 2018Co-Authors: Victor Veitch, Morgane Austern, Wenda Zhou, David M Blei, Peter OrbanzAbstract:Empirical Risk Minimization is the main tool for prediction problems, but its extension to relational data remains unsolved. We solve this problem using recent ideas from graph sampling theory to (i) define an empirical Risk for relational data and (ii) obtain stochastic gradients for this empirical Risk that are automatically unbiased. This is achieved by considering the method by which data is sampled from a graph as an explicit component of model design. By integrating fast implementations of graph sampling schemes with standard automatic differentiation tools, we provide an efficient turnkey solver for the Risk Minimization problem. We establish basic theoretical properties of the procedure. Finally, we demonstrate relational ERM with application to two non-standard problems: one-stage training for semi-supervised node classification, and learning embedding vectors for vertex attributes. Experiments confirm that the turnkey inference procedure is effective in practice, and that the sampling scheme used for model specification has a strong effect on model performance. Code is available at this https URL.
Meredith Y. Smith - One of the best experts on this subject based on the ideXlab platform.
-
Quality of Reporting on the Evaluation of Risk Minimization Programs: A Systematic Review
Drug Safety, 2020Co-Authors: Andrea M. Russell, Elaine H. Morrato, Rebecca M. Lovett, Meredith Y. SmithAbstract:Introduction Risk Minimization programs are interventions mandated by regulatory agencies to ensure that benefits of pharmaceutical products outweigh Risks. Many regulatory agencies require programs be evaluated for effectiveness; however, the quality of evidence has limited the ability to definitively determine if programs improve drug safety. Objective The aim of this systematic review was to assess and describe the current status of reporting on the effectiveness of pharmaceutical Risk management programs. Methods Peer-reviewed articles published between January 2012 and December 2018 were selected from three online databases (MEDLINE, PubMed, Embase). Eligible studies reported on effectiveness evaluations of mandated Risk Minimization measures (beyond labeling) and were written in English. Two reviewers independently examined 2744 titles of articles and 52 full articles were included. Forty-eight sources of gray literature from conference abstract presentations and publicly available regulatory documents were also included. Results Key opportunities for improvement in reporting included the provision of information regarding (1) selection, design, and testing of Risk Minimization measures, (2) implementation of programs, (3) process and outcome metrics, including the extent to which programs reached the intended audience, were integrated into the target healthcare settings, or were sustained over time, and (4) burden of the program on the healthcare system and implications for patient access. Conclusions Gaps in reporting of Risk Minimization program evaluation studies were identified. Addressing gaps will help build the evidence base regarding Risk Minimization initiatives, as well as ensure that programs are maximally effective and minimally burdensome on the healthcare system, and do not unduly interfere with patient access to the medicine.
-
Quality of Reporting on the Evaluation of Risk Minimization Programs: A Systematic Review
Drug safety, 2020Co-Authors: Andrea M. Russell, Elaine H. Morrato, Rebecca M. Lovett, Meredith Y. SmithAbstract:Risk Minimization programs are interventions mandated by regulatory agencies to ensure that benefits of pharmaceutical products outweigh Risks. Many regulatory agencies require programs be evaluated for effectiveness; however, the quality of evidence has limited the ability to definitively determine if programs improve drug safety. The aim of this systematic review was to assess and describe the current status of reporting on the effectiveness of pharmaceutical Risk management programs. Peer-reviewed articles published between January 2012 and December 2018 were selected from three online databases (MEDLINE, PubMed, Embase). Eligible studies reported on effectiveness evaluations of mandated Risk Minimization measures (beyond labeling) and were written in English. Two reviewers independently examined 2744 titles of articles and 52 full articles were included. Forty-eight sources of gray literature from conference abstract presentations and publicly available regulatory documents were also included. Key opportunities for improvement in reporting included the provision of information regarding (1) selection, design, and testing of Risk Minimization measures, (2) implementation of programs, (3) process and outcome metrics, including the extent to which programs reached the intended audience, were integrated into the target healthcare settings, or were sustained over time, and (4) burden of the program on the healthcare system and implications for patient access. Gaps in reporting of Risk Minimization program evaluation studies were identified. Addressing gaps will help build the evidence base regarding Risk Minimization initiatives, as well as ensure that programs are maximally effective and minimally burdensome on the healthcare system, and do not unduly interfere with patient access to the medicine.
-
Advancing the Field of Pharmaceutical Risk Minimization Through Application of Implementation Science Best Practices
Drug Safety, 2014Co-Authors: Meredith Y. Smith, Elaine MorratoAbstract:Regulators are increasingly mandating the use of pharmaceutical Risk-Minimization programs for a variety of medicinal products. To date, however, evaluations of these programs have shown mixed results and relatively little attention has been directed at diagnosing the specific factors contributing to program success or lack thereof. Given the growing use of these programs in many different patient populations, it is imperative to understand how best to design, deliver, disseminate, and assess them. In this paper, we argue that current approaches to designing, implementing, and evaluating Risk-Minimization programs could be improved by applying evidence- and theory-based ‘best practices’ from implementation science. We highlight commonly encountered challenges and gaps in the design, implementation, and evaluation of pharmaceutical Risk-Minimization initiatives and propose three key recommendations to address these issues: (1) Risk-Minimization program design should utilize models and frameworks that guide what should be done to produce successful outcomes and what questions should be addressed to evaluate program success; (2) intervention activities and tools should be theoretically grounded and evidence based; and (3) evaluation plans should incorporate a mixed-methods approach, pragmatic trial designs, and a range of outcomes. Regulators, practitioners, policy makers, and researchers are encouraged to apply these best practices in order to improve the public health impact of this important regulatory tool.
Sandro Ridella - One of the best experts on this subject based on the ideXlab platform.
-
structural Risk Minimization and rademacher complexity for regression
The European Symposium on Artificial Neural Networks, 2012Co-Authors: Davide Anguita, Alessandro Ghio, Luca Oneto, Sandro RidellaAbstract:The Structural Risk Minimization principle allows estimating the generalization ability of a learned hypothesis by measuring the com- plexity of the entire hypothesis class. Two of the most recent and effective complexity measures are the Rademacher Complexity and the Maximal Discrepancy, which have been applied to the derivation of generalization bounds for kernel classifiers. In this work, we extend their application to the regression framework.
Leon Bottou - One of the best experts on this subject based on the ideXlab platform.
-
Invariant Risk Minimization
arXiv: Machine Learning, 2019Co-Authors: Martin Arjovsky, Leon Bottou, Ishaan Gulrajani, David Lopez-pazAbstract:We introduce Invariant Risk Minimization (IRM), a learning paradigm to estimate invariant correlations across multiple training distributions. To achieve this goal, IRM learns a data representation such that the optimal classifier, on top of that data representation, matches for all training distributions. Through theory and experiments, we show how the invariances learned by IRM relate to the causal structures governing the data and enable out-of-distribution generalization.
-
vicinal Risk Minimization
Neural Information Processing Systems, 2000Co-Authors: Olivier Chapelle, Leon Bottou, Jason Weston, Vladimir VapnikAbstract:The Vicinal Risk Minimization principle establishes a bridge between generative models and methods derived from the Structural Risk Minimization Principle such as Support Vector Machines or Statistical Regularization. We explain how VRM provides a framework which integrates a number of existing algorithms, such as Parzen windows, Support Vector Machines, Ridge Regression, Constrained Logistic Classifiers and Tangent-Prop. We then show how the approach implies new algorithms for solving problems usually associated with generative models. New algorithms are described for dealing with pattern recognition problems with very different pattern distributions and dealing with unlabeled data. Preliminary empirical results are presented.
-
NIPS - Vicinal Risk Minimization
2000Co-Authors: Olivier Chapelle, Leon Bottou, Jason Weston, Vladimir VapnikAbstract:The Vicinal Risk Minimization principle establishes a bridge between generative models and methods derived from the Structural Risk Minimization Principle such as Support Vector Machines or Statistical Regularization. We explain how VRM provides a framework which integrates a number of existing algorithms, such as Parzen windows, Support Vector Machines, Ridge Regression, Constrained Logistic Classifiers and Tangent-Prop. We then show how the approach implies new algorithms for solving problems usually associated with generative models. New algorithms are described for dealing with pattern recognition problems with very different pattern distributions and dealing with unlabeled data. Preliminary empirical results are presented.
-
structural Risk Minimization for character recognition
Neural Information Processing Systems, 1991Co-Authors: Isabelle Guyon, Vladimir Vapnik, Bernhard E Boser, Leon Bottou, Sara A SollaAbstract:The method of Structural Risk Minimization refers to tuning the capacity of the classifier to the available amount of training data. This capacity is influenced by several factors, including: (1) properties of the input space, (2) nature and structure of the classifier, and (3) learning algorithm. Actions based on these three factors are combined here to control the capacity of linear classifiers and improve generalization on the problem of handwritten digit recognition.
-
NIPS - Structural Risk Minimization for Character Recognition
1991Co-Authors: Isabelle Guyon, Vladimir Vapnik, Bernhard E Boser, Leon Bottou, Sara A SollaAbstract:The method of Structural Risk Minimization refers to tuning the capacity of the classifier to the available amount of training data. This capacity is influenced by several factors, including: (1) properties of the input space, (2) nature and structure of the classifier, and (3) learning algorithm. Actions based on these three factors are combined here to control the capacity of linear classifiers and improve generalization on the problem of handwritten digit recognition.
