Much has been written about theory and practice in the law, and the tension between practitioners and theorists. Judges do not cite theoretical articles often; they rarely "apply" theories to particular cases. These arguments are not revisited. Instead the Essay explores the working and interaction of theory and practice, practitioners and theorists. The Essay starts with a story about solving a legal issue using our intellectual tools - theory, practice, and their progenies: experience and "gut." Next the Essay elaborates on the nature of theory, practice, experience and "gut." The third part of the Essay discusses theories that are helpful to practitioners and those that are less helpful. The Essay concludes that practitioners theorize, and theorists practice. They use these intellectual tools differently because the goals and orientations of theorists and practitioners, and the constraints under which they act, differ. Theory, practice, experience and "gut" help us think, remember, decide and create. They complement each other like the two sides of the same coin: distinct but inseparable.

Of Theory and Practice

Multi-Task Learning (MTL) is a learning paradigm in machine learning and its aim is to leverage useful information contained in multiple related tasks to help improve the generalization performance of all the tasks. In this paper, we give a survey for MTL from the perspective of algorithmic modeling, applications and theoretical analyses. For algorithmic modeling, we give a definition of MTL and then classify different MTL algorithms into five categories, including feature learning approach, low-rank approach, task clustering approach, task relation learning approach and decomposition approach as well as discussing the characteristics of each approach. In order to improve the performance of learning tasks further, MTL can be combined with other learning paradigms including semi-supervised learning, active learning, unsupervised learning, reinforcement learning, multi-view learning and graphical models. When the number of tasks is large or the data dimensionality is high, we review online, parallel and distributed MTL models as well as dimensionality reduction and feature hashing to reveal their computational and storage advantages. Many real-world applications use MTL to boost their performance and we review representative works in this paper. Finally, we present theoretical analyses and discuss several future directions for MTL.

A Survey on Multi-Task Learning

Given the potential risk of X-ray radiation to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. Currently, the main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction algorithms, but they need to access raw data, whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, deconvolution network, and shortcut connections into the residual encoder–decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection.

/pdf/low-dose-ct-with-a-residual-encoder-decoder-convolutional-2mk77wlfg8.pdf

Low-Dose CT With a Residual Encoder-Decoder Convolutional Neural Network

A deep convolutional neural network model to classify heartbeats

Alzheimer's disease (AD) is associated with neurodegeneration in vulnerable limbic and heteromodal regions of the cerebral cortex, detectable in vivo using magnetic resonance imaging. It is not clear whether abnormalities of cortical anatomy in AD can be reliably measured across different subject samples, how closely they track symptoms, and whether they are detectable prior to symptoms. An exploratory map of cortical thinning in mild AD was used to define regions of interest that were applied in a hypothesis-driven fashion to other subject samples. Results demonstrate a reliably quantifiable in vivo signature of abnormal cortical anatomy in AD, which parallels known regional vulnerability to AD neuropathology. Thinning in vulnerable cortical regions relates to symptom severity even in the earliest stages of clinical symptoms. Furthermore, subtle thinning is present in asymptomatic older controls with brain amyloid binding as detected with amyloid imaging. The reliability and clinical validity of AD-related cortical thinning suggests potential utility as an imaging biomarker. This “disease signature” approach to cortical morphometry, in which disease effects are mapped across the cortical mantle and then used to define ROIs for hypothesis-driven analyses, may provide a powerful methodological framework for studies of neuropsychiatric diseases.

The Cortical Signature of Alzheimer's Disease: Regionally Specific Cortical Thinning Relates to Symptom Severity in Very Mild to Mild AD Dementia and is Detectable in Asymptomatic Amyloid-Positive Individuals

The electrocardiogram (ECG) is one of the most important techniques for heart disease diagnosis. Many traditional methodologies of feature extraction and classification have been widely applied to ECG analysis. However, the effectiveness and efficiency of such methodologies remain to be improved, and much existing research did not consider the separation of training and testing samples from the same set of patients (so called inter-patient scheme). To cope with these issues, in this paper, we propose a method to classify ECG signals using wavelet packet entropy (WPE) and random forests (RF) following the Association for the Advancement of Medical Instrumentation (AAMI) recommendations and the inter-patient scheme. Specifically, we firstly decompose the ECG signals by wavelet packet decomposition (WPD), and then calculate entropy from the decomposed coefficients as representative features, and finally use RF to build an ECG classification model. To the best of our knowledge, it is the first time that WPE and RF are used to classify ECG following the AAMI recommendations and the inter-patient scheme. Extensive experiments are conducted on the publicly available MIT–BIH Arrhythmia database and influence of mother wavelets and level of decomposition for WPD, type of entropy and the number of base learners in RF on the performance are also discussed. The experimental results are superior to those by several state-of-the-art competing methods, showing that WPE and RF is promising for ECG classification.

https://www.mdpi.com/1099-4300/18/8/285/pdf

ECG Classification Using Wavelet Packet Entropy and Random Forests

Crude oil is one of the most important types of energy for the global economy, and hence it is very attractive to understand the movement of crude oil prices. However, the sequences of crude oil prices usually show some characteristics of nonstationarity and nonlinearity, making it very challenging for accurate forecasting crude oil prices. To cope with this issue, in this paper, we propose a novel approach that integrates complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and extreme gradient boosting (XGBOOST), so-called CEEMDAN-XGBOOST, for forecasting crude oil prices. Firstly, we use CEEMDAN to decompose the nonstationary and nonlinear sequences of crude oil prices into several intrinsic mode functions (IMFs) and one residue. Secondly, XGBOOST is used to predict each IMF and the residue individually. Finally, the corresponding prediction results of each IMF and the residue are aggregated as the final forecasting results. To demonstrate the performance of the proposed approach, we conduct extensive experiments on the West Texas Intermediate (WTI) crude oil prices. The experimental results show that the proposed CEEMDAN-XGBOOST outperforms some state-of-the-art models in terms of several evaluation metrics.

/pdf/a-ceemdan-and-xgboost-based-approach-to-forecast-crude-oil-1h93md4m8v.pdf

A CEEMDAN and XGBOOST-Based Approach to Forecast Crude Oil Prices

MPPCEDE: Multi-population parallel co-evolutionary differential evolution for parameter optimization

Image encryption is one of the essential tasks in image security. In this paper, we propose a novel approach that integrates a hyperchaotic system, pixel-level Dynamic Filtering, DNA computing, and operations on 3D Latin Cubes, namely DFDLC, for image encryption. Specifically, the approach consists of five stages: (1) a newly proposed 5D hyperchaotic system with two positive Lyapunov exponents is applied to generate a pseudorandom sequence; (2) for each pixel in an image, a filtering operation with different templates called dynamic filtering is conducted to diffuse the image; (3) DNA encoding is applied to the diffused image and then the DNA-level image is transformed into several 3D DNA-level cubes; (4) Latin cube is operated on each DNA-level cube; and (5) all the DNA cubes are integrated and decoded to a 2D cipher image. Extensive experiments are conducted on public testing images, and the results show that the proposed DFDLC can achieve state-of-the-art results in terms of several evaluation criteria.

Image Encryption Based on Pixel-Level Diffusion with Dynamic Filtering and DNA-Level Permutation with 3D Latin Cubes.

Alzheimer's disease (AD) is the most common form of dementia that causes progressive impairment of memory and other cognitive functions. Multivariate regression models have been studied in AD for revealing relationships between neuroimaging measures and cognitive scores to understand how structural changes in brain can influence cognitive status. Existing regression methods, however, do not explicitly model dependence relation among multiple scores derived from a single cognitive test. It has been found that such dependence can deteriorate the performance of these methods. To overcome this limitation, we propose an efficient sparse Bayesian multi-task learning algorithm, which adaptively learns and exploits the dependence to achieve improved prediction performance. The proposed algorithm is applied to a real world neuroimaging study in AD to predict cognitive performance using MRI scans. The effectiveness of the proposed algorithm is demonstrated by its superior prediction performance over multiple state-of-the-art competing methods and accurate identification of compact sets of cognition-relevant imaging biomarkers that are consistent with prior knowledge.

http://dsp.ucsd.edu/~zhilin/papers/Zhang_CVPR2012.pdf

Taiyong Li

Papers

ECG Classification Using Wavelet Packet Entropy and Random Forests

A CEEMDAN and XGBOOST-Based Approach to Forecast Crude Oil Prices

MPPCEDE: Multi-population parallel co-evolutionary differential evolution for parameter optimization

Image Encryption Based on Pixel-Level Diffusion with Dynamic Filtering and DNA-Level Permutation with 3D Latin Cubes.

Sparse Bayesian multi-task learning for predicting cognitive outcomes from neuroimaging measures in Alzheimer's disease