2016 年的期刊文章 · 第 32 页

共 1345 篇,按发表时间排,新的在前,每页 40 篇。这一页是 2016 到 2016。

按期刊、按主题看 →

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 页

Econometrica Econometrics & Economics · 发表于 2016

Farms, Families, and Markets: New Evidence on Completeness of Markets in Agricultural Settings

Daniel LaFave, Duncan Thomas

摘要

The farm household model has played a central role in improving the understanding of small-scale agricultural households and non-farm enterprises. Under the assumptions that all current and future markets exist and that farmers treat all prices as given, the model simplifies households' simultaneous production and consumption decisions into a recursive form in which production can be treated as independent of preferences of household members. These assumptions, which are the foundation of a large literature in labor and development, have been tested and not rejected in several important studies including Benjamin (1992). Using multiple waves of longitudinal survey data from Central Java, Indonesia, this paper tests a key prediction of the recursive model: demand for farm labor is unrelated to the demographic composition of the farm household. The prediction is unambiguously rejected. The rejection cannot be explained by contamination due to unobserved heterogeneity that is fixed at the farm level, local area shocks or farm-specific shocks that affect changes in household composition and farm labor demand. We conclude that the recursive form of the farm household model is not consistent with the data. Developing empirically tractable models of farm households when markets are incomplete remains an important challenge. *(abstract via PubMed)*

JMLR Other · 发表于 2016

Extremal Mechanisms for Local Differential Privacy

Peter Kairouz, Sewoong Oh, Pramod Viswanath

摘要

Local differential privacy has recently surfaced as a strong measure of privacy in contexts where personal information remains private even from data analysts. Working in a setting where both the data providers and data analysts want to maximize the utility of statistical analyses performed on the released data, we study the fundamental trade-off between local differential privacy and utility. This trade-off is formulated as a constrained optimization problem: maximize utility subject to local differential privacy constraints. We introduce a combinatorial family of extremal privatization mechanisms, which we call staircase mechanisms, and show that it contains the optimal privatization mechanisms for a broad class of information theoretic utilities such as mutual information and $f$-divergences. We further prove that for any utility function and any privacy level, solving the privacy-utility maximization problem is equivalent to solving a finite-dimensional linear program, the outcome of which is the optimal staircase mechanism. However, solving this linear program can be computationally expensive since it has a number of variables that is exponential in the size of the alphabet the data lives in. To account for this, we show that two simple privatization mechanisms, the binary and randomized response mechanisms, are universally optimal in the low and high privacy regimes, and well approximate the intermediate regime.

JMLR Clinical Trials · 发表于 2016

Extracting PICO Sentences from Clinical Trial Reports using Supervised Distant Supervision

Byron C. Wallace, Joël Kuiper, Aakash Sharma, Mingxi Zhu, Iain J. Marshall

摘要

Systematic reviews underpin Evidence Based Medicine (EBM) by addressing precise clinical questions via comprehensive synthesis of all relevant published evidence. Authors of systematic reviews typically define a Population/Problem, Intervention, Comparator, and Outcome (a PICO criteria) of interest, and then retrieve, appraise and synthesize results from all reports of clinical trials that meet these criteria. Identifying PICO elements in the full-texts of trial reports is thus a critical yet time-consuming step in the systematic review process. We seek to expedite evidence synthesis by developing machine learning models to automatically extract sentences from articles relevant to PICO elements. Collecting a large corpus of training data for this task would be prohibitively expensive. Therefore, we derive distant supervision (DS) with which to train models using previously conducted reviews. DS entails heuristically deriving 'soft' labels from an available structured resource. However, we have access only to unstructured, free-text summaries of PICO elements for corresponding articles; we must derive from these the desired sentence-level annotations. To this end, we propose a novel method -- supervised distant supervision (SDS) -- that uses a small amount of direct supervision to better exploit a large corpus of distantly labeled instances by learning to pseudo-annotate articles using the available DS. We show that this approach tends to outperform existing methods with respect to automated PICO extraction.

JMLR Optimization & Learning Theory · 发表于 2016

Exploration of the (Non-)Asymptotic Bias and Variance of Stochastic Gradient Langevin Dynamics

Sebastian J. Vollmer, Konstantinos C. Zygalakis, Yee Whye Teh

摘要

Applying standard Markov chain Monte Carlo (MCMC) algorithms to large data sets is computationally infeasible. The recently proposed stochastic gradient Langevin dynamics (SGLD) method circumvents this problem in three ways: it generates proposed moves using only a subset of the data, it skips the Metropolis- Hastings accept-reject step, and it uses sequences of decreasing step sizes. In Teh et al. (2014), we provided the mathematical foundations for the decreasing step size SGLD, including consistency and a central limit theorem. However, in practice the SGLD is run for a relatively small number of iterations, and its step size is not decreased to zero. The present article investigates the behaviour of the SGLD with fixed step size. In particular we characterise the asymptotic bias explicitly, along with its dependence on the step size and the variance of the stochastic gradient. On that basis a modified SGLD which removes the asymptotic bias due to the variance of the stochastic gradients up to first order in the step size is derived. Moreover, we are able to obtain bounds on the finite-time bias, variance and mean squared error (MSE). The theory is illustrated with a Gaussian toy model for which the bias and the MSE for the estimation of moments can be obtained explicitly. For this toy model we study the gain of the SGLD over the standard Euler method in the limit of large data sets.

JMLR Networks & Graphs · 发表于 2016

Exact Inference on Gaussian Graphical Models of Arbitrary Topology using Path-Sums

P.-L. Giscard, Z. Choo, S. J. Thwaite, D. Jaksch

摘要

We present the path-sum formulation for exact statistical inference of marginals on Gaussian graphical models of arbitrary topology. The path-sum formulation gives the covariance between each pair of variables as a branched continued fraction of finite depth and breadth. Our method originates from the closed- form resummation of infinite families of terms of the walk-sum representation of the covariance matrix. We prove that the path- sum formulation always exists for models whose covariance matrix is positive definite: i.e. it is valid for both walk-summable and non-walk-summable graphical models of arbitrary topology. We show that for graphical models on trees the path-sum formulation is equivalent to Gaussian belief propagation. We also recover, as a corollary, an existing result that uses determinants to calculate the covariance matrix. We show that the path-sum formulation formulation is valid for arbitrary partitions of the inverse covariance matrix. We give detailed examples demonstrating our results.

JMLR Other · 发表于 2016

Estimation from Pairwise Comparisons: Sharp Minimax Bounds with Topology Dependence

Nihar B. Shah, Sivaraman Balakrishnan, Joseph Bradley, Abhay Parekh, Kannan Ramchandran, Martin J. Wainwright

摘要

Data in the form of pairwise comparisons arises in many domains, including preference elicitation, sporting competitions, and peer grading among others. We consider parametric ordinal models for such pairwise comparison data involving a latent vector $w^* \in \mathbb{R}^d$ that represents the “qualities” of the $d$ items being compared; this class of models includes the two most widely used parametric models---the Bradley-Terry-Luce (BTL) and the Thurstone models. Working within a standard minimax framework, we provide tight upper and lower bounds on the optimal error in estimating the quality score vector $w^*$ under this class of models. The bounds depend on the topology of the comparison graph induced by the subset of pairs being compared, via the spectrum of the Laplacian of the comparison graph. Thus, in settings where the subset of pairs may be chosen, our results provide principled guidelines for making this choice. Finally, we compare these error rates to those under cardinal measurement models and show that the error rates in the ordinal and cardinal settings have identical scalings apart from constant pre- factors.

JMLR Networks & Graphs Optimization & Learning Theory · 发表于 2016

Estimating Diffusion Networks: Recovery Conditions, Sample Complexity and Soft-thresholding Algorithm

Manuel Gomez-Rodriguez, Le Song, Hadi Daneshm, Bernhard Schölkopf

摘要

Information spreads across social and technological networks, but often the network structures are hidden from us and we only observe the traces left by the diffusion processes, called cascades. Can we recover the hidden network structures from these observed cascades? What kind of cascades and how many cascades do we need? Are there some network structures which are more difficult than others to recover? Can we design efficient inference algorithms with provable guarantees? Despite the increasing availability of cascade data and methods for inferring networks from these data, a thorough theoretical understanding of the above questions remains largely unexplored in the literature. In this paper, we investigate the network structure inference problem for a general family of continuous- time diffusion models using an $\ell_1$-regularized likelihood maximization framework. We show that, as long as the cascade sampling process satisfies a natural incoherence condition, our framework can recover the correct network structure with high probability if we observe $O(d^3 \log N)$ cascades, where $d$ is the maximum number of parents of a node and $N$ is the total number of nodes. Moreover, we develop a simple and efficient soft-thresholding network inference algorithm which demonstrate the match between our theoretical prediction and empirical results. In practice, this new algorithm also outperforms other alternatives in terms of the accuracy of recovering hidden diffusion networks.

JMLR Causal Inference · 发表于 2016

Estimating Causal Structure Using Conditional DAG Models

Chris. J. Oates, Jim Q. Smith, Sach Mukherjee

摘要

This paper considers inference of causal structure in a class of graphical models called conditional DAGs. These are directed acyclic graph (DAG) models with two kinds of variables, primary and secondary. The secondary variables are used to aid in the estimation of the structure of causal relationships between the primary variables. We prove that, under certain assumptions, such causal structure is identifiable from the joint observational distribution of the primary and secondary variables. We give causal semantics for the model class, put forward a score-based approach for estimation and establish consistency results. Empirical results demonstrate gains compared with formulations that treat all variables on an equal footing, or that ignore secondary variables. The methodology is motivated by applications in biology that involve multiple data types and is illustrated here using simulated data and in an analysis of molecular data from the Cancer Genome Atlas.

JMLR High-Dimensional Networks & Graphs · 发表于 2016

Equivalence of Graphical Lasso and Thresholding for Sparse Graphs

Somayeh Sojoudi

摘要

This paper is concerned with the problem of finding a sparse graph capturing the conditional dependence between the entries of a Gaussian random vector, where the only available information is a sample correlation matrix. A popular approach to address this problem is the graphical lasso technique, which employs a sparsity-promoting regularization term. This paper derives a simple condition under which the computationally- expensive graphical lasso behaves the same as the simple heuristic method of thresholding. This condition depends only on the solution of graphical lasso and makes no direct use of the sample correlation matrix or the regularization coefficient. It is proved that this condition is always satisfied if the solution of graphical lasso is close to its first-order Taylor approximation or equivalently the regularization term is relatively large. This condition is tested on several random problems, and it is shown that graphical lasso and the thresholding method lead to highly similar results in the case where a sparse graph is sought. We also conduct two case studies on brain connectivity networks of twenty subjects based on fMRI data and the topology identification of electrical circuits to support the findings of this work on the similarity of graphical lasso and thresholding.

Econometrica Econometrics & Economics · 发表于 2016

Equilibrium Labor Turnover, Firm Growth, and Unemployment

Melvyn G. Coles, Dale T. Mortensen

摘要

This paper considers equilibrium quit turnover in a frictional labor market with costly hiring by firms, where large firms employ many workers and face both aggregate and firm specific productivity shocks.There is exogenous firm turnover as new (small) startups enter the market over time, while some existing firms fail and exit.Individual firm growth rates are disperse and evolve stochastically.The paper highlights how dynamic monopsony, where firms trade off lower wages against higher (endogenous) employee quit rates, yields excessive job-to-job quits.Such quits directly crowd out the reemployment prospects of the unemployed.With finite firm productivity states, stochastic equilibrium is fully tractable and can be computed using standard numerical techniques. *(abstract via OpenAlex)*

JMLR Other · 发表于 2016

End-to-End Training of Deep Visuomotor Policies

Sergey Levine, Chelsea Finn, Trevor Darrell, Pieter Abbeel

摘要

Policy search methods can allow robots to learn control policies for a wide range of tasks, but practical applications of policy search often require hand-engineered components for perception, state estimation, and low-level control. In this paper, we aim to answer the following question: does training the perception and control systems jointly end-to-end provide better performance than training each component separately? To this end, we develop a method that can be used to learn policies that map raw image observations directly to torques at the robot's motors. The policies are represented by deep convolutional neural networks (CNNs) with 92,000 parameters, and are trained using a guided policy search method, which transforms policy search into supervised learning, with supervision provided by a simple trajectory-centric reinforcement learning method. We evaluate our method on a range of real-world manipulation tasks that require close coordination between vision and control, such as screwing a cap onto a bottle, and present simulated comparisons to a range of prior policy search methods.

Econometrica Other · 发表于 2016

Elite Recruitment and Political Stability: The Impact of the Abolition of China's Civil Service Exam

Ying Bai, Ruixue Jia

摘要

This paper studies how the abolition of an elite recruitment system—China's civil exam system that lasted over 1,300 years—affects political stability. Employing a panel data set across 262 prefectures and exploring the variations in the quotas on the entry-level exam candidates, we find that higher quotas per capita were associated with a higher probability of revolution participation after the abolition and a higher incidence of uprisings in 1911 that marked the end of the 2,000 years of imperial rule. This finding is robust to various checks including using the number of small rivers and short-run exam performance before the quota system as instruments. The patterns in the data appear most consistent with the interpretation that in regions with higher quotas per capita under the exam system, more would-be elites were negatively affected by the abolition. In addition, we document that modern human capital in the form of those studying in Japan also contributed to the revolution and that social capital strengthened the effect of quotas on revolution participation. *(abstract via OpenAlex)*

JMLR Latent Variables & Mixtures · 发表于 2016

Electronic Health Record Analysis via Deep Poisson Factor Models

Ricardo Henao, James T. Lu, Joseph E. Lucas, Jeffrey Ferranti, Lawrence Carin

摘要

Electronic Health Record (EHR) phenotyping utilizes patient data captured through normal medical practice, to identify features that may represent computational medical phenotypes. These features may be used to identify at-risk patients and improve prediction of patient morbidity and mortality. We present a novel deep multi-modality architecture for EHR analysis (applicable to joint analysis of multiple forms of EHR data), based on Poisson Factor Analysis (PFA) modules. Each modality, composed of observed counts, is represented as a Poisson distribution, parameterized in terms of hidden binary units. Information from different modalities is shared via a deep hierarchy of common hidden units. Activation of these binary units occurs with probability characterized as Bernoulli- Poisson link functions, instead of more traditional logistic link functions. In addition, we demonstrate that PFA modules can be adapted to discriminative modalities. To compute model parameters, we derive efficient Markov Chain Monte Carlo (MCMC) inference that scales efficiently, with significant computational gains when compared to related models based on logistic link functions. To explore the utility of these models, we apply them to a subset of patients from the Duke-Durham patient cohort. We identified a cohort of over 16,000 patients with Type 2 Diabetes Mellitus (T2DM) based on diagnosis codes and laboratory tests out of our patient population of over 240,000. Examining the common hidden units uniting the PFA modules, we identify patient features that represent medical concepts. Experiments indicate that our learned features are better able to predict mortality and morbidity than clinical features identified previously in a large-scale clinical trial.

JMLR Spatial & Time Series · 发表于 2016

Efficient Computation of Gaussian Process Regression for Large Spatial Data Sets by Patching Local Gaussian Processes

Chiwoo Park, Jianhua Z. Huang

摘要

This paper develops an efficient computational method for solving a Gaussian process (GP) regression for large spatial data sets using a collection of suitably defined local GP regressions. The conventional local GP approach first partitions a domain into multiple non-overlapping local regions, and then fits an independent GP regression for each local region using the training data belonging to the region. Two key issues with the local GP are (1) the prediction around the boundary of a local region is not as accurate as the prediction at interior of the local region, and (2) two local GP regressions for two neighboring local regions produce different predictions at the boundary of the two regions, creating undesirable discontinuity in the prediction. We address these issues by constraining the predictions of local GP regressions sharing a common boundary to satisfy the same boundary constraints, which in turn are estimated by the data. The boundary constrained local GP regressions are solved by a finite element method. Our approach shows competitive performance when compared with several state- of-the-art methods using two synthetic data sets and three real data sets.

JMLR Other · 发表于 2016

ERRATA: On the Estimation of the Gradient Lines of a Density and the Consistency of the Mean-Shift Algorithm

Ery Arias-Castro, David Mason, Bruno Pelletier

摘要

ERRATA to the paper On the Estimation of the Gradient Lines of a Density and the Consistency of the Mean-Shift Algorithm.

JMLR Bayesian Methods Machine Learning Econometrics & Economics · 发表于 2016

Dual Control for Approximate Bayesian Reinforcement Learning

Edgar D. Klenske, Philipp Hennig

摘要

Control of non-episodic, finite-horizon dynamical systems with uncertain dynamics poses a tough and elementary case of the exploration-exploitation trade-off. Bayesian reinforcement learning, reasoning about the effect of actions and future observations, offers a principled solution, but is intractable. We review, then extend an old approximate approach from control theory---where the problem is known as dual control---in the context of modern regression methods, specifically generalized linear regression. Experiments on simulated systems show that this framework offers a useful approximation to the intractable aspects of Bayesian RL, producing structured exploration strategies that differ from standard RL approaches. We provide simple examples for the use of this framework in (approximate) Gaussian process regression and feedforward neural networks for the control of exploration.

JMLR Econometrics & Economics · 发表于 2016

Double or Nothing: Multiplicative Incentive Mechanisms for Crowdsourcing

Nihar B. Shah, Dengyong Zhou

摘要

Crowdsourcing has gained immense popularity in machine learning applications for obtaining large amounts of labeled data. Crowdsourcing is cheap and fast, but suffers from the problem of low-quality data. To address this fundamental challenge in crowdsourcing, we propose a simple payment mechanism to incentivize workers to answer only the questions that they are sure of and skip the rest. We show that surprisingly, under a mild and natural no-free-lunch requirement, this mechanism is the one and only incentive-compatible payment mechanism possible. We also show that among all possible incentive- compatible mechanisms (that may or may not satisfy no-free- lunch), our mechanism makes the smallest possible payment to spammers. We further extend our results to a more general setting in which workers are required to provide a quantized confidence for each question. Interestingly, this unique mechanism takes a multiplicative form. The simplicity of the mechanism is an added benefit. In preliminary experiments involving over 900 worker-task pairs, we observe a significant drop in the error rates under this unique mechanism for the same or lower monetary expenditure.

JMLR Machine Learning Networks & Graphs · 发表于 2016

Domain-Adversarial Training of Neural Networks

Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, Mario March, Victor Lempitsky

摘要

We introduce a new representation learning approach for domain adaptation, in which data at training and test time come from similar but different distributions. Our approach is directly inspired by the theory on domain adaptation suggesting that, for effective domain transfer to be achieved, predictions must be made based on features that cannot discriminate between the training (source) and test (target) domains.

JMLR Other · 发表于 2016

Distribution-Matching Embedding for Visual Domain Adaptation

Mahsa Baktashmotlagh, Mehrtash Harandi, Mathieu Salzmann

摘要

Domain-invariant representations are key to addressing the domain shift problem where the training and test examples follow different distributions. Existing techniques that have attempted to match the distributions of the source and target domains typically compare these distributions in the original feature space. This space, however, may not be directly suitable for such a comparison, since some of the features may have been distorted by the domain shift, or may be domain specific. In this paper, we introduce a Distribution-Matching Embedding approach: An unsupervised domain adaptation method that overcomes this issue by mapping the data to a latent space where the distance between the empirical distributions of the source and target examples is minimized. In other words, we seek to extract the information that is invariant across the source and target data. In particular, we study two different distances to compare the source and target distributions: the Maximum Mean Discrepancy and the Hellinger distance. Furthermore, we show that our approach allows us to learn either a linear embedding, or a nonlinear one. We demonstrate the benefits of our approach on the tasks of visual object recognition, text categorization, and WiFi localization.

JMLR Meta-Analysis & Data Integration Optimization & Learning Theory · 发表于 2016

Distributed Submodular Maximization

Baharan Mirzasoleiman, Amin Karbasi, Rik Sarkar, Andreas Krause

摘要

Many large-scale machine learning problems--clustering, non- parametric learning, kernel machines, etc.--require selecting a small yet representative subset from a large dataset. Such problems can often be reduced to maximizing a submodular set function subject to various constraints. Classical approaches to submodular optimization require centralized access to the full dataset, which is impractical for truly large-scale problems. In this paper, we consider the problem of submodular function maximization in a distributed fashion. We develop a simple, two- stage protocol GREEDI, that is easily implemented using MapReduce style computations. We theoretically analyze our approach, and show that under certain natural conditions, performance close to the centralized approach can be achieved. We begin with monotone submodular maximization subject to a cardinality constraint, and then extend this approach to obtain approximation guarantees for (not necessarily monotone) submodular maximization subject to more general constraints including matroid or knapsack constraints. In our extensive experiments, we demonstrate the effectiveness of our approach on several applications, including sparse Gaussian process inference and exemplar based clustering on tens of millions of examples using Hadoop.

JMLR Meta-Analysis & Data Integration · 发表于 2016

Distributed Coordinate Descent Method for Learning with Big Data

Peter Richtárik, Martin Takáč

摘要

In this paper we develop and analyze Hydra: HYbriD cooRdinAte descent method for solving loss minimization problems with big data. We initially partition the coordinates (features) and assign each partition to a different node of a cluster. At every iteration, each node picks a random subset of the coordinates from those it owns, independently from the other computers, and in parallel computes and applies updates to the selected coordinates based on a simple closed-form formula. We give bounds on the number of iterations sufficient to approximately solve the problem with high probability, and show how it depends on the data and on the partitioning. We perform numerical experiments with a LASSO instance described by a 3TB matrix.

JMLR Causal Inference · 发表于 2016

Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks

Joris M. Mooij, Jonas Peters, Dominik Janzing, Jakob Zscheischler, Bernhard Schölkopf

摘要

The discovery of causal relationships from purely observational data is a fundamental problem in science. The most elementary form of such a causal discovery problem is to decide whether $X$ causes $Y$ or, alternatively, $Y$ causes $X$, given joint observations of two variables $X,Y$. An example is to decide whether altitude causes temperature, or vice versa, given only joint measurements of both variables. Even under the simplifying assumptions of no confounding, no feedback loops, and no selection bias, such bivariate causal discovery problems are challenging. Nevertheless, several approaches for addressing those problems have been proposed in recent years. We review two families of such methods: methods based on Additive Noise Models (ANMs) and Information Geometric Causal Inference (IGCI). We present the benchmark CauseEffectPairs that consists of data for 100 different cause-effect pairs selected from 37 data sets from various domains (e.g., meteorology, biology, medicine, engineering, economy, etc.) and motivate our decisions regarding the ground truth causal directions of all pairs. We evaluate the performance of several bivariate causal discovery methods on these real-world benchmark data and in addition on artificially simulated data. Our empirical results on real-world data indicate that certain methods are indeed able to distinguish cause from effect using only purely observational data, although more benchmark data would be needed to obtain statistically significant conclusions. One of the best performing methods overall is the method based on Additive Noise Models that has originally been proposed by Hoyer et al. (2009), which obtains an accuracy of 63 $\pm$ 10 % and an AUC of 0.74 $\pm$ 0.05 on the real-world benchmark. As the main theoretical contribution of this work we prove the consistency of that method.

JMLR Other · 发表于 2016

Dimension-free Concentration Bounds on Hankel Matrices for Spectral Learning

François Denis, Mattias Gybels, Amaury Habrard

摘要

Learning probabilistic models over strings is an important issue for many applications. Spectral methods propose elegant solutions to the problem of inferring weighted automata from finite samples of variable-length strings drawn from an unknown target distribution $p$. These methods rely on a singular value decomposition of a matrix $\v{H}_S$, called the empirical Hankel matrix, that records the frequencies of (some of) the observed strings $S$. The accuracy of the learned distribution depends both on the quantity of information embedded in $\v{H}_S$ and on the distance between $\v{H}_S$ and its mean $\v{H}_p$. Existing concentration bounds seem to indicate that the concentration over $\v{H}_p$ gets looser with its dimensions, suggesting that it might be necessary to bound the dimensions of $\v{H}_S$ for learning. We prove new dimension-free concentration bounds for classical Hankel matrices and several variants, based on prefixes or factors of strings, that are useful for learning. Experiments demonstrate that these bounds are tight and that they significantly improve existing (dimension-dependent) bounds. One consequence of these results is that the spectral learning approach remains consistent even if all the observations are recorded within the empirical matrix.

JMLR Other · 发表于 2016

Differentially Private Data Releasing for Smooth Queries

Ziteng Wang, Chi Jin, Kai Fan, Jiaqi Zhang, Junliang Huang, Yiqiao Zhong, Liwei Wang

摘要

In the past few years, differential privacy has become a standard concept in the area of privacy. One of the most important problems in this field is to answer queries while preserving differential privacy. In spite of extensive studies, most existing work on differentially private query answering assumes the data are discrete (i.e., in $\{0,1\}^d$) and focuses on queries induced by \emph{Boolean} functions. In real applications however, continuous data are at least as common as binary data. Thus, in this work we explore a less studied topic, namely, differential privately query answering for continuous data with continuous function. As a first step towards the continuous case, we study a natural class of linear queries on continuous data which we refer to as smooth queries. A linear query is said to be $K$-smooth if it is specified by a function defined on $[-1,1]^d$ whose partial derivatives up to order $K$ are all bounded. We develop two $\epsilon$-differentially private mechanisms which are able to answer all smooth queries. The first mechanism outputs a summary of the database and can then give answers to the queries. The second mechanism is an improvement of the first one and it outputs a synthetic database. The two mechanisms both achieve an accuracy of $O (n^{-\frac{K}{2d+K}}/\epsilon )$. Here we assume that the dimension $d$ is a constant. It turns out that even in this parameter setting (which is almost trivial in the discrete case), using existing discrete mechanisms to answer the smooth queries is difficult and requires more noise. Our mechanisms are based on $L_{\infty}$-approximation of (transformed) smooth functions by low-degree even trigonometric polynomials with uniformly bounded coefficients. We also develop practically efficient variants of the mechanisms with promising experimental results.

SiM Survival Analysis · 发表于 2016

Developing points‐based risk‐scoring systems in the presence of competing risks

Peter C. Austin, Douglas S. Lee, Ralph B. D'Agostino, Jason P. Fine

摘要

Predicting the occurrence of an adverse event over time is an important issue in clinical medicine. Clinical prediction models and associated points‐based risk‐scoring systems are popular statistical methods for summarizing the relationship between a multivariable set of patient risk factors and the risk of the occurrence of an adverse event. Points‐based risk‐scoring systems are popular amongst physicians as they permit a rapid assessment of patient risk without the use of computers or other electronic devices. The use of such points‐based risk‐scoring systems facilitates evidence‐based clinical decision making. There is a growing interest in cause‐specific mortality and in non‐fatal outcomes. However, when considering these types of outcomes, one must account for competing risks whose occurrence precludes the occurrence of the event of interest. We describe how points‐based risk‐scoring systems can be developed in the presence of competing events. We illustrate the application of these methods by developing risk‐scoring systems for predicting cardiovascular mortality in patients hospitalized with acute myocardial infarction. Code in the R statistical programming language is provided for the implementation of the described methods. © 2016 The Authors. Statistics in Medicine

JMLR Other · 发表于 2016

Decrypting “Cryptogenic” Epilepsy: Semi-supervised Hierarchical Conditional Random Fields For Detecting Cortical Lesions In MRI-Negative Patients

Bilal Ahmed, Thomas Thesen, Karen E. Blackmon, Ruben Kuzniekcy, Orrin Devinsky, Carla E. Brodley

摘要

Focal cortical dysplasia (FCD) is the most common cause of pediatric epilepsy and the third most common cause in adults with treatment-resistant epilepsy. Surgical resection of the lesion is the most effective treatment to stop seizures. Technical advances in MRI have revolutionized the diagnosis of FCD, leading to high success rates for resective surgery. However, 45% of histologically confirmed FCD patients have normal MRIs (MRI-negative). Without a visible lesion, the success rate of surgery drops from 66% to 29%. In this work, we cast the problem of detecting potential FCD lesions using MRI scans of MRI-negative patients in an image segmentation framework based on hierarchical conditional random fields (HCRF). We use surface based morphometry to model the cortical surface as a two-dimensional surface which is then segmented at multiple scales to extract superpixels of different sizes. Each superpixel is assigned an outlier score by comparing it to a control population. The lesion is detected by fusing the outlier probabilities across multiple scales using a tree- structured HCRF. The proposed method achieves a higher detection rate, with superior recall and precision on a sample of twenty MRI-negative FCD patients as compared to a baseline across four morphological features and their combinations.

JMLR Other · 发表于 2016

Data-driven Rank Breaking for Efficient Rank Aggregation

Ashish Khetan, Sewoong Oh

摘要

Rank aggregation systems collect ordinal preferences from individuals to produce a global ranking that represents the social preference. Rank-breaking is a common practice to reduce the computational complexity of learning the global ranking. The individual preferences are broken into pairwise comparisons and applied to efficient algorithms tailored for independent paired comparisons. However, due to the ignored dependencies in the data, naive rank-breaking approaches can result in inconsistent estimates. The key idea to produce accurate and consistent estimates is to treat the pairwise comparisons unequally, depending on the topology of the collected data. In this paper, we provide the optimal rank-breaking estimator, which not only achieves consistency but also achieves the best error bound. This allows us to characterize the fundamental tradeoff between accuracy and complexity. Further, the analysis identifies how the accuracy depends on the spectral gap of a corresponding comparison graph.

JMLR Other · 发表于 2016

DSA: Decentralized Double Stochastic Averaging Gradient Algorithm

Aryan Mokhtari, Alejandro Ribeiro

摘要

This paper considers optimization problems where nodes of a network have access to summands of a global objective. Each of these local objectives is further assumed to be an average of a finite set of functions. The motivation for this setup is to solve large scale machine learning problems where elements of the training set are distributed to multiple computational elements. The decentralized double stochastic averaging gradient (DSA) algorithm is proposed as a solution alternative that relies on: (i) The use of local stochastic averaging gradients. (ii) Determination of descent steps as differences of consecutive stochastic averaging gradients. Strong convexity of local functions and Lipschitz continuity of local gradients is shown to guarantee linear convergence of the sequence generated by DSA in expectation. Local iterates are further shown to approach the optimal argument for almost all realizations. The expected linear convergence of DSA is in contrast to the sublinear rate characteristic of existing methods for decentralized stochastic optimization. Numerical experiments on a logistic regression problem illustrate reductions in convergence time and number of feature vectors processed until convergence relative to these other alternatives.

JMLR Bayesian Methods Nonparametric & Semiparametric Latent Variables & Mixtures · 发表于 2016

CrossCat: A Fully Bayesian Nonparametric Method for Analyzing Heterogeneous, High Dimensional Data

Vikash Mansinghka, Patrick Shafto, Eric Jonas, Cap Petschulat, Max Gasner, Joshua B. Tenenbaum

摘要

There is a widespread need for statistical methods that can analyze high-dimensional datasets without imposing restrictive or opaque modeling assumptions. This paper describes a domain- general data analysis method called CrossCat. CrossCat infers multiple non-overlapping views of the data, each consisting of a subset of the variables, and uses a separate nonparametric mixture to model each view. CrossCat is based on approximately Bayesian inference in a hierarchical, nonparametric model for data tables. This model consists of a Dirichlet process mixture over the columns of a data table in which each mixture component is itself an independent Dirichlet process mixture over the rows; the inner mixture components are simple parametric models whose form depends on the types of data in the table. CrossCat combines strengths of mixture modeling and Bayesian network structure learning. Like mixture modeling, CrossCat can model a broad class of distributions by positing latent variables, and produces representations that can be efficiently conditioned and sampled from for prediction. Like Bayesian networks, CrossCat represents the dependencies and independencies between variables, and thus remains accurate when there are multiple statistical signals. Inference is done via a scalable Gibbs sampling scheme; this paper shows that it works well in practice. This paper also includes empirical results on heterogeneous tabular data of up to 10 million cells, such as hospital cost and quality measures, voting records, unemployment rates, gene expression measurements, and images of handwritten digits. CrossCat infers structure that is consistent with accepted findings and common-sense knowledge in multiple domains and yields predictive accuracy competitive with generative, discriminative, and model-free alternatives.

JMLR Longitudinal Data · 发表于 2016

Cross-Corpora Unsupervised Learning of Trajectories in Autism Spectrum Disorders

Huseyin Melih Elibol, Vincent Nguyen, Scott Linderman, Matthew Johnson, Amna Hashmi, Finale Doshi-Velez

摘要

Patients with developmental disorders, such as autism spectrum disorder (ASD), present with symptoms that change with time even if the named diagnosis remains fixed. For example, language impairments may present as delayed speech in a toddler and difficulty reading in a school-age child. Characterizing these trajectories is important for early treatment. However, deriving these trajectories from observational sources is challenging: electronic health records only reflect observations of patients at irregular intervals and only record what factors are clinically relevant at the time of observation. Meanwhile, caretakers discuss daily developments and concerns on social media.

Econometrica Testing & Uncertainty Econometrics & Economics · 发表于 2016

Credibility of Confidence Sets in Nonstandard Econometric Problems

Ulrich K. Müller, Andriy Norets

摘要

Confidence intervals are commonly used to describe parameter uncertainty. In nonstandard problems, however, their frequentist coverage property does not guarantee that they do so in a reasonable fashion. For instance, confidence intervals may be empty or extremely short with positive probability, even if they are based on inverting powerful tests. We apply a betting framework to formalize the “reasonableness ” of confidence intervals as descriptions of parameter uncertainty, and use it for two purposes. First, we quantify the degree of unreasonableness of previously suggested confidence intervals in nonstandard problems. Second, we derive alternative confidence sets that are reasonable by construction. We apply our framework to inference about a parameter near a boundary and a local-to-unity autoregressive root. We find that previously suggested confidence intervals are not reasonable, and numerically determine alternative confidence sets that satisfy our criteria. JEL classification: C18 *(abstract via OpenAlex)*

JMLR Functional Data Latent Variables & Mixtures · 发表于 2016

Covariance-based Clustering in Multivariate and Functional Data Analysis

Francesca Ieva, Anna Maria Paganoni, Nicholas Tarabelloni

摘要

In this paper we propose a new algorithm to perform clustering of multivariate and functional data. We study the case of two populations different in their covariances, rather than in their means. The algorithm relies on a proper quantification of distance between the estimated covariance operators of the populations, and subdivides data in two groups maximising the distance between their induced covariances. The naive implementation of such an algorithm is computationally forbidding, so we propose a heuristic formulation with a much lighter complexity and we study its convergence properties, along with its computational cost. We also propose to use an enhanced estimator for the estimation of discrete covariances of functional data, namely a linear shrinkage estimator, in order to improve the precision of the clustering. We establish the effectiveness of our algorithm through applications to both synthetic data and a real data set coming from a biomedical context, showing also how the use of shrinkage estimation may lead to substantially better results.

JMLR Optimization & Learning Theory · 发表于 2016

Convex Regression with Interpretable Sharp Partitions

Ashley Petersen, Noah Simon, Daniela Witten

摘要

We consider the problem of predicting an outcome variable on the basis of a small number of covariates, using an interpretable yet non-additive model. We propose convex regression with interpretable sharp partitions (CRISP) for this task. CRISP partitions the covariate space into blocks in a data- adaptive way, and fits a mean model within each block. Unlike other partitioning methods, CRISP is fit using a non-greedy approach by solving a convex optimization problem, resulting in low- variance fits. We explore the properties of CRISP, and evaluate its performance in a simulation study and on a housing price data set.

JMLR Diagnostics & Prediction · 发表于 2016

Convex Calibration Dimension for Multiclass Loss Matrices

Harish G. Ramaswamy, Shivani Agarwal

摘要

We study consistency properties of surrogate loss functions for general multiclass learning problems, defined by a general multiclass loss matrix. We extend the notion of classification calibration, which has been studied for binary and multiclass 0-1 classification problems (and for certain other specific learning problems), to the general multiclass setting, and derive necessary and sufficient conditions for a surrogate loss to be calibrated with respect to a loss matrix in this setting. We then introduce the notion of convex calibration dimension of a multiclass loss matrix, which measures the smallest "size" of a prediction space in which it is possible to design a convex surrogate that is calibrated with respect to the loss matrix. We derive both upper and lower bounds on this quantity, and use these results to analyze various loss matrices. In particular, we apply our framework to study various subset ranking losses, and use the convex calibration dimension as a tool to show both the existence and non-existence of various types of convex calibrated surrogates for these losses. Our results strengthen recent results of Duchi et al. (2010) and Calauzènes et al. (2012) on the non-existence of certain types of convex calibrated surrogates in subset ranking. We anticipate the convex calibration dimension may prove to be a useful tool in the study and design of surrogate losses for general multiclass learning problems.

JMLR Other · 发表于 2016

Convergence of an Alternating Maximization Procedure

Andreas Andresen, Vladimir Spokoiny

摘要

We derive two convergence results for a sequential alternating maximization procedure to approximate the maximizer of random functionals such as the realized log likelihood in MLE estimation. We manage to show that the sequence attains the same deviation properties as shown for the profile M-estimator by Andresen and Spokoiny (2013), that means a finite sample Wilks and Fisher theorem. Further under slightly stronger smoothness constraints on the random functional we can show nearly linear convergence to the global maximizer if the starting point for the procedure is well chosen.

JMLR Causal Inference · 发表于 2016

Control Function Instrumental Variable Estimation of Nonlinear Causal Effect Models

Zijian Guo, Dylan S. Small

摘要

The instrumental variable method consistently estimates the effect of a treatment when there is unmeasured confounding and a valid instrumental variable. A valid instrumental variable is a variable that is independent of unmeasured confounders and affects the treatment but does not have a direct effect on the outcome beyond its effect on the treatment. Two commonly used estimators for using an instrumental variable to estimate a treatment effect are the two stage least squares estimator and the control function estimator. For linear causal effect models, these two estimators are equivalent, but for nonlinear causal effect models, the estimators are different. We provide a systematic comparison of these two estimators for nonlinear causal effect models and develop an approach to combing the two estimators that generally performs better than either one alone. We show that the control function estimator is a two stage least squares estimator with an augmented set of instrumental variables. If these augmented instrumental variables are valid, then the control function estimator can be much more efficient than usual two stage least squares without the augmented instrumental variables while if the augmented instrumental variables are not valid, then the control function estimator may be inconsistent while the usual two stage least squares remains consistent. We apply the Hausman test to test whether the augmented instrumental variables are valid and construct a pretest estimator based on this test. The pretest estimator is shown to work well in a simulation study. An application to the effect of exposure to violence on time preference is considered.

Econometrica Other · 发表于 2016

Consumption Commitments and Habit Formation

Raj Chetty, Adam Szeidl

摘要

We analyze the implications of household-level adjustment costs for the dynamics of aggregate consumption. We show that an economy in which agents have “consumption commitments” is approximately equivalent to a habit formation model in which the habit stock is a weighted average of past consumption if idiosyncratic risk is large relative to aggregate risk. Consumption commitments can thus explain the empirical regularity that consumption is excessively sensitive and excessively smooth, findings that are typically attributed to habit formation. Unlike habit formation and other theories, but consistent with empirical evidence, the consumption commitments model also predicts that excess sensitivity and smoothness vanish for large shocks. These results suggest that behavior previously attributed to habit formation may be better explained by adjustment costs. We develop additional testable predictions to further distinguish the commitment and habit models and show that the two models have different welfare implications. *(abstract via OpenAlex)*

Econometrica Other · 发表于 2016

Consistent Probabilistic Social Choice

Florian Brandl, Felix Brandt, Hans Georg Seedig

摘要

Two fundamental axioms in social choice theory are consistency with respect to a variable electorate and consistency with respect to components of similar alternatives. In the context of traditional non-probabilistic social choice, these axioms are incompatible with each other. We show that in the context of probabilistic social choice, these axioms uniquely characterize a function proposed by Fishburn (Rev. Econ. Stud., 51(4), 683--692, 1984). Fishburn's function returns so-called maximal lotteries, i.e., lotteries that correspond to optimal mixed strategies of the underlying plurality game. Maximal lotteries are guaranteed to exist due to von Neumann's Minimax Theorem, are almost always unique, and can be efficiently computed using linear programming. *(abstract via OpenAlex)*

JMLR Other · 发表于 2016

Consistent Distribution-Free $K$-Sample and Independence Tests for Univariate Random Variables

Ruth Heller, Yair Heller, Shachar Kaufman, Barak Brill, Malka Gorfine

摘要

A popular approach for testing if two univariate random variables are statistically independent consists of partitioning the sample space into bins, and evaluating a test statistic on the binned data. The partition size matters, and the optimal partition size is data dependent. While for detecting simple relationships coarse partitions may be best, for detecting complex relationships a great gain in power can be achieved by considering finer partitions. We suggest novel consistent distribution-free tests that are based on summation or maximization aggregation of scores over all partitions of a fixed size. We show that our test statistics based on summation can serve as good estimators of the mutual information. Moreover, we suggest regularized tests that aggregate over all partition sizes, and prove those are consistent too. We provide polynomial-time algorithms, which are critical for computing the suggested test statistics efficiently. We show that the power of the regularized tests is excellent compared to existing tests, and almost as powerful as the tests based on the optimal (yet unknown in practice) partition size, in simulations as well as on a real data example.

JMLR Latent Variables & Mixtures Spatial & Time Series · 发表于 2016

Consistent Algorithms for Clustering Time Series

Azadeh Khaleghi, Daniil Ryabko, Jérémie Mary, Philippe Preux

摘要

The problem of clustering is considered for the case where every point is a time series. The time series are either given in one batch (offline setting), or they are allowed to grow with time and new time series can be added along the way (online setting). We propose a natural notion of consistency for this problem, and show that there are simple, computationally efficient algorithms that are asymptotically consistent under extremely weak assumptions on the distributions that generate the data. The notion of consistency is as follows. A clustering algorithm is called consistent if it places two time series into the same cluster if and only if the distribution that generates them is the same. In the considered framework the time series are allowed to be highly dependent, and the dependence can have arbitrary form. If the number of clusters is known, the only assumption we make is that the (marginal) distribution of each time series is stationary ergodic. No parametric, memory or mixing assumptions are made. When the number of clusters is unknown, stronger assumptions are provably necessary, but it is still possible to devise nonparametric algorithms that are consistent under very general conditions. The theoretical findings of this work are illustrated with experiments on both synthetic and real data.

第 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 页