Landmark annotation for training images is essential for many learning tasks in computer vision, such as object detection, tracking, and alignment. Image annotation is typically conducted manually, which is both labor-intensive and error-prone. To improve this process, this paper proposes a new approach to estimating the locations of a set of landmarks for a large image ensemble using manually ...

Given an ensemble of N × N random matrices with independent entries chosen from a nice probability distribution, a natural question is whether the empirical spectral measures of typical matrices converge to some limiting measure as N → ∞. It has been shown that the limiting spectral distribution for the ensemble of real symmetric matrices is a semi-circle, and that the distribution for real sym...

The Adaptive Boosting (AdaBoost) classifier is a widely used ensemble learning framework, and it can get good classification results on general datasets. However, challenging to apply the AdaBoost directly pulmonary nodule detection of labeled unlabeled lung CT images since there are still some drawbacks method. Therefore, solve data problem, semi-supervised using an improved sparrow search alg...

Using longer spectra we reanalyze spectral properties of the two-body random ensemble studied 30 years ago. At the center of the spectra the old results are largely confirmed, and we show that the nonergodicity is essentially due to the variance of the lowest moments of the spectra. The longer spectra allow us to test and reach the limits of validity of French's correction for the number varian...

We investigate the theoretical links between a regression ensemble and a linearly combined classification ensemble. First, we reformulate the Tumer & Ghosh model for linear combiners in a regression context; we then exploit this new formulation to generalise the concept of the “Ambiguity decomposition”, previously defined only for regression tasks, to classification problems. Finally, we propos...

Dimensionality reduction is a commonly used tool in machine learning, especially when dealing with high dimensional data. We consider semi-supervised graph based dimensionality reduction in this paper, and a novel dimensionality reduction algorithm called constrained Laplacian Eigenmap (CLE) is proposed. Suppose the data set contains r classes, and for each class we have some labeled points. CL...

Prediction of seminal quality with statistical learning tools is an emerging methodology in decision support systems in biomedical engineering and is very useful in early diagnosis of seminal patients and selection of semen donors candidates. However, as is common in medical diagnosis, seminal quality prediction faces the class imbalance problem. In this paper, we propose a novel supervised ens...

The most common machine learning approach is supervised learning, which uses labeled data for building predictive models. However, in many practical problems, the availability of annotated data is limited due to the expensive, tedious and time-consuming annotation procedure. At the same, unlabeled data can be easily available in large amounts. This is especially pronounced for predictive modell...