WorldCat Identities

Girosi, Federico

Overview
Works: 38 works in 86 publications in 1 language and 1,500 library holdings
Genres: Conference proceedings 
Roles: Editor
Classifications: R864, 651.504261
Publication Timeline
Key
Publications about  Federico Girosi Publications about Federico Girosi
Publications by  Federico Girosi Publications by Federico Girosi
Most widely held works by Federico Girosi
Extrapolating evidence of health information technology savings and costs by Federico Girosi ( )
10 editions published in 2005 in English and held by 1,055 WorldCat member libraries worldwide
In 2003, RAND Health began a broad study to better understand the role and importance of Electronic Medical Record Systems (EMR-S) in improving health and reducing healthcare costs, and to help inform government actions that could maximize EMR-S benefits and increase its use. This report provides the technical details and results of one component of that study: national-level efficiency savings brought about by using Healthcare Information Technology (HIT). We quantify those savings--what results from the ability to perform the same task with fewer resources (money, time, personnel, etc.)-- by providing a methodological framework to scale empirical evidence on the effect of HIT to the national level and to project it into the future. A key element of this framework is a projection of the rates of adoption of HIT in the inpatient setting and in the ambulatory/outpatient setting. Next, from the evidence found in our search of peer-reviewed and gray literature (the body of reports and studies produced by local government agencies, private organizations, and educational facilities that have not been reviewed and published in journals or other standard research publications), we considered savings from 10 different sources (5 inpatient; 5 outpatient). Then, we compared the efficiency savings with the costs the nation has to incur in order to be able to realize those savings, using a modeling framework analogous to the one developed for the extrapolation of savings and cost data from the literature or given to us by providers. We found that savings outweigh costs by a factor of 5, which implies that, even if a large portion of savings is not realized, the ratio of benefit to cost is still larger than 1. Finally, we studied what might be the effect of those financial incentives presented to providers that lower the cost of EMR-S and quicken the pace of HIT adoption. A general result that does not depend on the size of the behavioral response of physicians is that incentive programs are more likely to be cost-effective if they start early and do not last long, but are sizable. The report concludes with a summary chapter. The report should be of interest to healthcare IT professionals, other healthcare executives and researchers, and officials in the government responsible for health policy
Demographic forecasting by Federico Girosi ( Book )
6 editions published in 2008 in English and held by 260 WorldCat member libraries worldwide
Neural networks for signal processing V : proceedings of the 1995 IEEE Workshop : fifth in a series of workshops organized by the IEEE Signal Processing Society Neural Networks Technical Committee by IEEE Workshop on Neural Networks for Signal Processing ( Book )
5 editions published in 1995 in English and held by 75 WorldCat member libraries worldwide
Employer self-insurance decisions and the implications of the Patient Protection and Affordable Care Act as modified by the Health Care and Education Reconciliation Act of 2010 (ACA) by Christine Eibner ( )
1 edition published in 2011 in English and held by 17 WorldCat member libraries worldwide
The Patient Protection and Affordable Care Act as amended by the Health Care and Education Reconciliation Act of 2010 (ACA) changes the regulatory environment within which health insurance policies on the small-group market are bought and sold. New regulations include rate bands that limit premium price variation, risk-adjustment policies that will transfer funds from low-actuarial-risk to high-actuarial-risk plans, and requirements that plans include "essential health benefits." While the new regulations will be applied to all non-grandfathered fully insured policies purchased by businesses with 100 or fewer workers, self-insured plans are exempt from these regulations. As a result, some firms may have a stronger incentive to offer self-insured plans after the ACA takes full effect. In this report we identify factors that influence employers' decisions to self-insure and estimate how the ACA will influence self-insurance rates. We also consider the implications of higher self-insurance rates for adverse selection in the non-self-insured small-group market and whether enrollees in self-insured plans receive different benefits than enrollees in fully-insured plans. Results are based on data analysis, literature review, findings from discussions with stakeholders, and microsimulation analysis using the COMPARE model. Overall, we find little evidence that self-insured plans differ systematically from fully insured plans in terms of benefit generosity, price, or claims denial rates. Stakeholders expressed significant concern about adverse selection in the health insurance exchanges due to regulatory exemptions for self-insured plans. However, our microsimulation analysis predicts a sizable increase in self-insurance only if comprehensive stop-loss policies become widely available after the ACA takes full effect and the expected cost of self-insuring with stop-loss is comparable to the cost of being fully insured in a market without rating regulations
Priors, stabilizers and basis functions: from regularization to radial, tensor and additive splines by Federico Girosi ( Book )
4 editions published in 1993 in English and held by 6 WorldCat member libraries worldwide
We had previously shown that regularization principles lead to approximation schemes, as Radial Basis Functions, which are equivalent to networks with one layer of hidden units, called Regularization Networks. In this paper we show that regularization networks encompass a much broader range of approximation schemes, including many of the popular general additive models, Breiman's hinge functions and some forms of Projection Pursuit Regression. In the probabilistic interpretation of regularization, the different classes of basis functions correspond to different classes of prior probabilities on the approximating function spaces, and therefore to different types of smoothness assumptions. In the final part of the paper, we also show a relation between activation functions of the Gaussian and sigmoidal type
A theory of networks for approximation and learning by Massachusetts Institute of Technology ( Book )
7 editions published between 1989 and 1994 in English and held by 6 WorldCat member libraries worldwide
Learning an input-output mapping from a set of examples, of the type that many neural networks have been constructed to perform, can be regarded as synthesizing an approximation of a multi-dimensional function, that is solving the problem of hypersurface reconstruction. From this point of view, this form of learning is closely related to classical approximation techniques. This paper considers the problems of an exact representation of the approximation of linear and nonlinear mappings in terms of simpler functions of fewer variables. Kolmogorov's theorem concerning the representation of functions of several variables in terms of functions of one variable turns out to be almost irrelevant in the context of networks for learning. We develop a theoretical framework for approximation based on regularization techniques that leads to a class of three-layer networks that we call Generalized Radial Basis Function (GRBF), since they are mathematically related to the well-known Radial Basis Functions, mainly used for strict interpolation tasks. GRBF networks are not only equivalent to generalized splines, but are also closely related to pattern recognition methods such as Parzen windows and potential functions and to several neural network algorithms, such as Kanerva's associative memory, backpropagation and Kohonen's topology preserving map. They also have an interesting interpretation in terms of prototypes that are synthesized and optimally combined during the learning stage. The paper introduces intriguing analogies with neurobiological data
Continuous stochastic cellular automata that have a stationary distribution and no detailed balance by Tomaso Poggio ( Book )
2 editions published in 1990 in English and held by 5 WorldCat member libraries worldwide
Abstract: "Marroquin and Ramirez (1990) have recently discovered a class of discrete stochastic cellular automata with Gibbsian invariant measure that have a non-reversible dynamic behavior. Practical applications include more powerful algorithms than the Metropolis algorithm to compute MRF models. In this paper we describe a large class of stochastic dynamical systems that has a Gibbs asymptotic distribution but does not satisfy reversibility. We characterize sufficient properties of a sub-class of stochastic differential equations in terms of the associated Fokker-Planck equation for the existence of an asymptotic probability distribution in the system of coordinates which is given. Practical implications include VLSI analog circuits to compute coupled MRF models."
Extensions of a theory of networks for approximation and learning : outliers and negative examples by Federico Girosi ( Book )
4 editions published in 1990 in English and held by 5 WorldCat member libraries worldwide
Learning an input output mapping from a set of examples, of the type that many neural networks have been constructed to perform, can be regarded as synthesizing an approximation of a multi dimensional function. From this point of view, this form of learning is closely related to regularization theory. The theory developed in Poggio and Girosi (1989) shows the equivalence between regularization and a class of three-layer networks that we call regularization networks or Hyper Basis Functions. These networks are not only equivalent to generalized splines, but are closely related to the classical Radial Basis Functions used for interpolation tasks and to several pattern recognition and neural network algorithms. In this note, we extend the theory by introducing ways of dealing with two aspects of learning: learning in the presence of unreliable examples and learning from positive and negative examples. These two extensions are interesting also from the point of view of the approximation of multivariate functions. The first extension corresponds to dealing with outliers among the sparse data. The second one corresponds to exploiting information about points or regions in the range of the function that are forbidden
Establishing state health insurance exchanges implications for health insurance enrollment, spending, and small businesses ( )
1 edition published in 2010 in English and held by 5 WorldCat member libraries worldwide
The RAND Corporation's Comprehensive Assessment of Reform Efforts microsimulation model was used to analyze the effects of the Patient Protection and Affordable Care Act (PPACA) on employers and enrollees in employer-sponsored health insurance, with a focus on small businesses and businesses offering coverage through health insurance exchanges. Outcomes assessed include the proportion of nonelderly Americans with insurance coverage, the number of employers offering health insurance, premium prices, total employer spending, and total government spending relative to what would have been observed without the policy change. The microsimulation predicts that PPACA will increase insurance offer rates among small businesses from 53 to 77 percent for firms with ten or fewer workers, from 71 to 90 percent for firms with 11 to 25 workers, and from 90 percent to nearly 100 percent for firms with 26 to 100 workers. Simultaneously, the uninsurance rate in the United States would fall from 19 to 6 percent of the nonelderly population. The increase in employer offer rates is driven by workers' demand for insurance, which increases due to an individual mandate requiring all people to obtain insurance policies. Employer penalties incentivizing businesses to offer coverage do not have a meaningful impact on outcomes. The model further predicts that approximately 60 percent of businesses will offer coverage through the health insurance exchanges after the reform. Under baseline assumptions, a total of 68 million people will enroll in the exchanges, of whom 35 million will receive exchange-based coverage from an employer
Some extensions of the k-means algorithm for image segmentation and pattern classification by J. L Marroquin ( Book )
2 editions published in 1993 in English and held by 4 WorldCat member libraries worldwide
The same state variables can be used to define algorithms for determining adaptively the optimal number of centers for clouds of data with space-varying density. Some examples of the application of these extensions are also given."
A connection between GRBF and MLP by Minoru Maruyama ( Book )
2 editions published between 1991 and 1992 in English and held by 4 WorldCat member libraries worldwide
Multiscale GRBF networks, on the other hand, can approximate MLP networks with a similar number of parameters."
Convergence rates of approximation by translates by Federico Girosi ( Book )
2 editions published in 1992 in English and held by 4 WorldCat member libraries worldwide
We give results both for approximation in the L₂ norm and in the L[subscript infinity] norm. The interesting feature of these results is that, thanks to the constructive nature of Jones' and Barron's lemma, an iterative procedure is defined that can achieve this rate."
Extensions of a Theory of Networks for Approximation and Learning: Dimensionality Reduction and Clustering by Massachusetts Institute of Technology ( Book )
6 editions published in 1990 in English and held by 3 WorldCat member libraries worldwide
Learning an input-output mapping from a set of examples, of the type that many neural networks have been constructed to perform, can be regarded as synthesizing an approximation of a multi-dimensional function. From this point of view, this form of learning is closely related to regularization theory. The theory developed in Poggio and Girosi (1989) shows the equivalence between regularization and a class of three-layer networks that we call regularization networks or Hyper Basis Functions. These networks are not only equivalent to generalized splines, but are also closely related to the classical Radial Basis Functions used for interpolation tasks and to several pattern recognition and neural network algorithms. In this note, we extend the theory by defining a general form of these networks with two sets of modifiable parameters in addition to the coefficients C sub alpha: moving centers and adjustable norm-weights. Moving the centers is equivalent to task-dependent clustering and changing the norm weights is equivalent to task-dependent dimensionality reduction. (KR)
The impact of the coverage - related provisions of the Patient Protection and Affordable Care Act on insurance coverage and state health care expenditures in Texas an analysis from Rand compare ( )
1 edition published in 2011 in English and held by 3 WorldCat member libraries worldwide
The Patient Protection and Affordable Care Act (ACA) contains substantial new requirements aimed at increasing rates of health insurance coverage. Because many of these provisions impose additional costs on the states, officials need reliable estimates of the likely impact of the ACA in their state. To demonstrate the usefulness of modeling for state-level decision making, RAND undertook a preliminary analysis of the impact of the ACA on five states -- California, Connecticut, Illinois, Montana, and Texas -- using the RAND COMPARE microsimulation model. For Texas, the model predicts that, in 2016 (the year that all of the provisions in the ACA related to coverage expansion will be fully implemented), the uninsured rate in Texas will fall to 6 percent; without the law, it would remain at 28 percent, the highest in the nation. The model projects that total state government spending on health care will be 10 percent higher for the combined 2011-2020 period because of the ACA
The impact of the coverage - related provisions of the Patient Protection and Affordable Care Act on insurance coverage and state health care expenditures in Montana an analysis from Rand compare ( )
1 edition published in 2011 in English and held by 3 WorldCat member libraries worldwide
The Patient Protection and Affordable Care Act (ACA) contains substantial new requirements aimed at increasing rates of health insurance coverage. Because many of these provisions impose additional costs on the states, officials need reliable estimates of the likely impact of the ACA in their state. To demonstrate the usefulness of modeling for state-level decisionmaking, RAND undertook a preliminary analysis of the impact of the ACA on five states -- California, Connecticut, Illinois, Montana, and Texas -- using the RAND COMPARE microsimulation model. For Montana, the model predicts that, in 2016 (the year that all of the provisions in the ACA related to coverage expansion will be fully implemented), the uninsured rate in Montana will fall to 3 percent; without the law, it would remain at 18 percent. The model projects that total state government spending on health care will be 3 percent higher for the combined 2011-2020 period because of the ACA
On the Relationship Between Generalization Error, Hypothesis Complexity, and Sample Complexity for Radial Basis Functions by Partha Niyogi ( Book )
3 editions published in 1994 in English and held by 3 WorldCat member libraries worldwide
In this paper, we bound the generalization error of a class of Radial Basis Function networks, for certain well defined function learning tasks, in terms of the number of parameters and number of examples. We show that the total generalization error is partly due to the insufficient representational capacity of the network (because of its finite size) and partly due to insufficient information about the target function (because of finite number of samples). We make several observations about generalization error which are valid irrespective of the approximation scheme. Our result also sheds light on ways to choose an appropriate network architecture for a particular problem
Estimating the global impact of improved diagnostic tools for the developing world ( Book )
1 edition published in 2007 in English and held by 3 WorldCat member libraries worldwide
This research brief summarizes research assessing how higher-quality and more-accessible clinical diagnostic tests could improve health outcomes in the developing world for a number of common diseases
Models of noise and robust estimates by Federico Girosi ( Book )
1 edition published in 1991 in English and held by 3 WorldCat member libraries worldwide
In this paper we show that, for a class of functions V, using these robust estimators corresponds to assuming that data are corrupted by Gaussian noise whose variance fluctuates according to some given probability distribution, that uniquely determines the shape of V."
The impact of the coverage - related provisions of the Patient Protection and Affordable Care Act on insurance coverage and state health care expenditures in Connecticut an analysis from Rand compare ( )
1 edition published in 2011 in English and held by 3 WorldCat member libraries worldwide
The Patient Protection and Affordable Care Act (ACA) contains substantial new requirements aimed at increasing rates of health insurance coverage. Because many of these provisions impose additional costs on the states, officials need reliable estimates of the likely impact of the ACA in their state. To demonstrate the usefulness of modeling for state-level decisionmaking, RAND undertook a preliminary analysis of the impact of the ACA on five states -- California, Connecticut, Illinois, Montana, and Texas -- using the RAND COMPARE microsimulation model. For Connecticut, the model predicts that, in 2016 (the year that all of the provisions in the ACA related to coverage expansion will be fully implemented), the uninsured rate in Connecticut will fall to 5 percent; without the law, it would remain at 11 percent. The model projects that total state government spending on health care will be 10 percent lower for the combined 2011-2020 period than it would be without the ACA, mostly because of federal subsidies for residents who would have been covered by Connecticut's state-run health insurance program (State-Administered General Assistance)
 
moreShow More Titles
fewerShow Fewer Titles
Audience Level
0
Audience Level
1
  Kids General Special  
Audience level: 0.61 (from 0.00 for Neural net ... to 0.93 for Neural net ...)
Languages
English (61)
Covers