SVM is a powerful technique and especially useful for data whose distribution is unknown (also known as non-regularity in data). Because the example considered here consisted of only two features, the SVM fitted by R here is also known as linear SVM. SVM is powered by a kernel for dealing with various kinds of data and its kernel can also be set during model tuning. Some such examples include gaussian and radial. Hence, SVM can also be used for non-linear data and does not require any assumptions about its functional form.

I am happy to introduce the package HCmodelSets, which is now available on CRAN. This package implements the methods proposed by Cox, D.R. and Battey, H.S. (2017). In particular it performs the reduction, exploratory and model selection phases given in the aforementioned reference. The software supports linear regression, likelihood-based fitting of generalized linear regression models and the proportional hazards model fitted by partial likelihood. The standard method described in the literature to deal with sparse regression is the LASSO proposed by Tibshirani (1996), which assumes sparsity of the effects.

Machine learning is a computer programming technique that uses statistical probabilities to give computers the ability to "learn" without being explicitly programmed. In essence, machine learning is getting computers to learn--and therefore act--they way humans do, improving their learning and knowledge over time autonomously. The idea is to get computers to act without being explicitly programmed. Machine learning utilizes development programs that can adjust when exposed to different external inputs. But the key to machine learning is inputting lots and lots of data into the student computer.

Machine learning is a computer programming technique that uses statistical probabilities to give computers the ability to "learn" without being explicitly programmed. In essence, machine learning is getting computers to learn--and therefore act--they way humans do, improving their learning and knowledge over time autonomously. The idea is to get computers to act without being explicitly programmed. Machine learning utilizes development programs that can adjust when exposed to different external inputs. But the key to machine learning is inputting lots and lots of data into the student computer.

When generating adversarial examples to attack deep neural networks (DNNs), $\ell_p$ norm of the added perturbation is usually used to measure the similarity between original image and adversarial example. However, such adversarial attacks may fail to capture key infomation hidden in the input. This work develops a more general attack model i.e., the structured attack that explores group sparsity in adversarial perturbations by sliding a mask through images aiming for extracting key structures. An ADMM (alternating direction method of multipliers)-based framework is proposed that can split the original problem into a sequence of analytically solvable subproblems and can be generalized to implement other state-of-the-art attacks. Strong group sparsity is achieved in adversarial perturbations even with the same level of distortion in terms of $\ell_p$ norm as the state-of-the-art attacks. Extensive experimental results on MNIST, CIFAR-10 and ImageNet show that our attack could be much stronger (in terms of smaller $\ell_0$ distortion) than the existing ones, and its better interpretability from group sparse structures aids in uncovering the origins of adversarial examples.

The high-dimensional data setting, in which p >> n, is a challenging statistical paradigm that appears in many real-world problems. In this setting, learning a compact, low-dimensional representation of the data can substantially help distinguish signal from noise. One way to achieve this goal is to perform subspace learning to estimate a small set of latent features that capture the majority of the variance in the original data. Most existing subspace learning models, such as PCA, assume that the data can be fully represented by its embedding in one or more latent subspaces. However, in this work, we argue that this assumption is not suitable for many high-dimensional datasets; often only some variables can easily be projected to a low-dimensional space. We propose a hybrid dimensionality reduction technique in which some features are mapped to a low-dimensional subspace while others remain in the original space. Our model leads to more accurate estimation of the latent space and lower reconstruction error. We present a simple optimization procedure for the resulting biconvex problem and show synthetic data results that demonstrate the advantages of our approach over existing methods. Finally, we demonstrate the effectiveness of this method for extracting meaningful features from both gene expression and video background subtraction datasets.

Self-taught learning is a technique that uses a large number of unlabeled data as source samples to improve the task performance on target samples. Compared with other transfer learning techniques, self-taught learning can be applied to a broader set of scenarios due to the loose restrictions on source data. However, knowledge transferred from source samples that are not sufficiently related to the target domain may negatively influence the target learner, which is referred to as negative transfer. In this paper, we propose a metric for the relevance between a source sample and target samples. To be more specific, both source and target samples are reconstructed through a single-layer autoencoder with a linear relationship between source samples and target samples simultaneously enforced. An l_{2,1}-norm sparsity constraint is imposed on the transformation matrix to identify source samples relevant to the target domain. Source domain samples that are deemed relevant are assigned pseudo-labels reflecting their relevance to target domain samples, and are combined with target samples in order to provide an expanded training set for classifier training. Local data structures are also preserved during source sample selection through spectral graph analysis. Promising results in extensive experiments show the advantages of the proposed approach.

As more and more devices rely on data sharing, the Internet of Things (IoT) is rapidly expanding. Developing strong privacy and security protocols for these systems is critical, as devices, applications, and communication networks become increasingly integrated. However, as these systems grow, cyber attacks on the IoT are becoming increasingly complex. These attacks often involve machine learning, earning themselves the name "smart attacks". The security solutions necessary to combat these attacks are computationally heavy and frequently involve a large communication load. Additionally, due to their relatively small computational abilities, many IoT devices are more vulnerable to attacks than computer systems.

In automatic speech processing systems, speaker diarization is a crucial front-end component to separate segments from different speakers. Inspired by the recent success of deep neural networks (DNNs) in semantic inferencing, triplet loss-based architectures have been successfully used for this problem. However, existing work utilizes conventional i-vectors as the input representation and builds simple fully connected networks for metric learning, thus not fully leveraging the modeling power of DNN architectures. This paper investigates the importance of learning effective representations from the sequences directly in metric learning pipelines for speaker diarization. More specifically, we propose to employ attention models to learn embeddings and the metric jointly in an end-to-end fashion. Experiments are conducted on the CALLHOME conversational speech corpus. The diarization results demonstrate that, besides providing a unified model, the proposed approach achieves improved performance when compared against existing approaches.

While it is believed that denoising is not always necessary in many big data applications, we show in this paper that denoising is helpful in urban traffic analysis by applying the method of bounded total variation denoising to the urban road traffic prediction and clustering problem. We propose two easy-to-implement methods to estimate the noise strength parameter in the denoising algorithm, and apply the denoising algorithm to GPS-based traffic data from Beijing taxi system. For the traffic prediction problem, we combine neural network and history matching method for roads randomly chosen from an urban area of Beijing. Numerical experiments show that the predicting accuracy is improved significantly by applying the proposed bounded total variation denoising algorithm. We also test the algorithm on clustering problem, where a recently developed clustering analysis method is applied to more than one hundred urban road segments in Beijing based on their velocity profiles. Better clustering result is obtained after denoising.