Employing a large dataset (at most, the order of n = 10^6), this study attempts enhance the literature on the comparison between regression and machine learning (ML)-based rent price prediction models by adding new empirical evidence and considering the spatial dependence of the observations. The regression-based approach incorporates the nearest neighbor Gaussian processes (NNGP) model, enabling the application of kriging to large datasets. In contrast, the ML-based approach utilizes typical models: extreme gradient boosting (XGBoost), random forest (RF), and deep neural network (DNN). The out-of-sample prediction accuracy of these models was compared using Japanese apartment rent data, with a varying order of sample sizes (i.e., n = 10^4, 10^5, 10^6). The results showed that, as the sample size increased, XGBoost and RF outperformed NNGP with higher out-of-sample prediction accuracy. XGBoost achieved the highest prediction accuracy for all sample sizes and error measures in both logarithmic and real scales and for all price bands (when n = 10^5 and 10^6). A comparison of several methods to account for the spatial dependence in RF showed that simply adding spatial coordinates to the explanatory variables may be sufficient.
相關內容
Classical asymptotic theory for statistical inference usually involves calibrating a statistic by fixing the dimension $d$ while letting the sample size $n$ increase to infinity. Recently, much effort has been dedicated towards understanding how these methods behave in high-dimensional settings, where $d_n$ and $n$ both increase to infinity together at some prescribed relative rate. This often leads to different inference procedures, depending on the assumptions about the dimensionality, leaving the practitioner in a bind: given a dataset with 100 samples in 20 dimensions, should they calibrate by assuming $n \gg d$, or $d_n/n \approx 0.2$? This paper considers the goal of dimension-agnostic inference -- developing methods whose validity does not depend on any assumption on $d_n$. We introduce a new, generic approach that uses variational representations of existing test statistics along with sample splitting and self-normalization to produce a new test statistic with a Gaussian limiting distribution. The resulting statistic can be viewed as a careful modification of degenerate U-statistics, dropping diagonal blocks and retaining off-diagonals. We exemplify our technique for a handful of classical problems including one-sample mean and covariance testing. Our tests are shown to have minimax rate-optimal power against appropriate local alternatives, and without explicitly targeting the high-dimensional setting their power is optimal up to a $\sqrt 2$ factor. A hidden advantage is that our proofs are simple and transparent. We end by describing several fruitful open directions.
Variational Bayes (VB) is a popular tool for Bayesian inference in statistical modeling. Recently, some VB algorithms are proposed to handle intractable likelihoods with applications such as approximate Bayesian computation. In this paper, we propose several unbiased estimators based on multilevel Monte Carlo (MLMC) for the gradient of Kullback-Leibler divergence between the posterior distribution and the variational distribution when the likelihood is intractable, but can be estimated unbiasedly. The new VB algorithm differs from the VB algorithms in the literature which usually render biased gradient estimators. Moreover, we incorporate randomized quasi-Monte Carlo (RQMC) sampling within the MLMC-based gradient estimators, which was known to provide a favorable rate of convergence in numerical integration. Theoretical guarantees for RQMC are provided in this new setting. Numerical experiments show that using RQMC in MLMC greatly speeds up the VB algorithm, and finds a better parameter value than some existing competitors do.
Today, most automated test generators, such as search-based software testing (SBST) techniques focus on achieving high code coverage. However, high code coverage is not sufficient to maximise the number of bugs found, especially when given a limited testing budget. In this paper, we propose an automated test generation technique that is also guided by the estimated degree of defectiveness of the source code. Parts of the code that are likely to be more defective receive more testing budget than the less defective parts. To measure the degree of defectiveness, we leverage Schwa, a notable defect prediction technique. We implement our approach into EvoSuite, a state of the art SBST tool for Java. Our experiments on the Defects4J benchmark demonstrate the improved efficiency of defect prediction guided test generation and confirm our hypothesis that spending more time budget on likely defective parts increases the number of bugs found in the same time budget.
This paper reviews two main types of prediction interval methods under a parametric framework. First, we describe methods based on an (approximate) pivotal quantity. Examples include the plug-in, pivotal, and calibration methods. Then we describe methods based on a predictive distribution (sometimes derived based on the likelihood). Examples include Bayesian, fiducial, and direct-bootstrap methods. Several examples involving continuous distributions along with simulation studies to evaluate coverage probability properties are provided. We provide specific connections among different prediction interval methods for the (log-)location-scale family of distributions. This paper also discusses general prediction interval methods for discrete data, using the binomial and Poisson distributions as examples. We also overview methods for dependent data, with application to time series, spatial data, and Markov random fields, for example.
Smoothing splines have been used pervasively in nonparametric regressions. However, the computational burden of smoothing splines is significant when the sample size $n$ is large. When the number of predictors $d\geq2$, the computational cost for smoothing splines is at the order of $O(n^3)$ using the standard approach. Many methods have been developed to approximate smoothing spline estimators by using $q$ basis functions instead of $n$ ones, resulting in a computational cost of the order $O(nq^2)$. These methods are called the basis selection methods. Despite algorithmic benefits, most of the basis selection methods require the assumption that the sample is uniformly-distributed on a hyper-cube. These methods may have deteriorating performance when such an assumption is not met. To overcome the obstacle, we develop an efficient algorithm that is adaptive to the unknown probability density function of the predictors. Theoretically, we show the proposed estimator has the same convergence rate as the full-basis estimator when $q$ is roughly at the order of $O[n^{2d/\{(pr+1)(d+2)\}}\quad]$, where $p\in[1,2]$ and $r\approx 4$ are some constants depend on the type of the spline. Numerical studies on various synthetic datasets demonstrate the superior performance of the proposed estimator in comparison with mainstream competitors.
Determination of posterior probability for go-no-go decision and predictive power are becoming increasingly common for resource optimization in clinical investigation. There are vast published literature on these topics; however, the terminologies are not consistently used across the literature. Further, there is a lack of consolidated presentation of various concepts of the probability of success. We attempted to fill this gap. This paper first provides a detailed derivation of these probability of success measures under the frequentist and Bayesian paradigms in a general setting. Subsequently, we have presented the analytical formula for these probability of success measures for continuous, binary, and time-to-event endpoints separately. This paper can be used as a single point reference to determine the following measures: (a) the conditional power (CP) based on interim results, (b) the predictive power of success (PPoS) based on interim results with or without prior distribution, and (d) the probability of success (PoS) for a prospective trial at the design stage. We have discussed both clinical success and trial success. This paper's discussion is mostly based on the normal approximation for prior distribution and the estimate of the parameter of interest. Besides, predictive power using the beta prior for the binomial case is also presented. Some examples are given for illustration. R functions to calculate CP and PPoS are available through the LongCART package. An R shiny app is also available at //ppos.herokuapp.com/.
There has been considerable growth and interest in industrial applications of machine learning (ML) in recent years. ML engineers, as a consequence, are in high demand across the industry, yet improving the efficiency of ML engineers remains a fundamental challenge. Automated machine learning (AutoML) has emerged as a way to save time and effort on repetitive tasks in ML pipelines, such as data pre-processing, feature engineering, model selection, hyperparameter optimization, and prediction result analysis. In this paper, we investigate the current state of AutoML tools aiming to automate these tasks. We conduct various evaluations of the tools on many datasets, in different data segments, to examine their performance, and compare their advantages and disadvantages on different test cases.
Accurately classifying malignancy of lesions detected in a screening scan plays a critical role in reducing false positives. Through extracting and analyzing a large numbers of quantitative image features, radiomics holds great potential to differentiate the malignant tumors from benign ones. Since not all radiomic features contribute to an effective classifying model, selecting an optimal feature subset is critical. This work proposes a new multi-objective based feature selection (MO-FS) algorithm that considers both sensitivity and specificity simultaneously as the objective functions during the feature selection. In MO-FS, we developed a modified entropy based termination criterion (METC) to stop the algorithm automatically rather than relying on a preset number of generations. We also designed a solution selection methodology for multi-objective learning using the evidential reasoning approach (SMOLER) to automatically select the optimal solution from the Pareto-optimal set. Furthermore, an adaptive mutation operation was developed to generate the mutation probability in MO-FS automatically. The MO-FS was evaluated for classifying lung nodule malignancy in low-dose CT and breast lesion malignancy in digital breast tomosynthesis. Compared with other commonly used feature selection methods, the experimental results for both lung nodule and breast lesion malignancy classification demonstrated that the feature set by selected MO-FS achieved better classification performance.
The word order between source and target languages significantly influences the translation quality in machine translation. Preordering can effectively address this problem. Previous preordering methods require a manual feature design, making language dependent design costly. In this paper, we propose a preordering method with a recursive neural network that learns features from raw inputs. Experiments show that the proposed method achieves comparable gain in translation quality to the state-of-the-art method but without a manual feature design.
Partial person re-identification (re-id) is a challenging problem, where only some partial observations (images) of persons are available for matching. However, few studies have offered a flexible solution of how to identify an arbitrary patch of a person image. In this paper, we propose a fast and accurate matching method to address this problem. The proposed method leverages Fully Convolutional Network (FCN) to generate certain-sized spatial feature maps such that pixel-level features are consistent. To match a pair of person images of different sizes, hence, a novel method called Deep Spatial feature Reconstruction (DSR) is further developed to avoid explicit alignment. Specifically, DSR exploits the reconstructing error from popular dictionary learning models to calculate the similarity between different spatial feature maps. In that way, we expect that the proposed FCN can decrease the similarity of coupled images from different persons and increase that of coupled images from the same person. Experimental results on two partial person datasets demonstrate the efficiency and effectiveness of the proposed method in comparison with several state-of-the-art partial person re-id approaches. Additionally, it achieves competitive results on a benchmark person dataset Market1501 with the Rank-1 accuracy being 83.58%.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191