Collaborating differently on different topics: a multi-relational approach to multi-task learning

Multi-task learning offers a way to benefit from synergy of multiple related prediction tasks via their joint modeling. Current multi-task techniques model related tasks jointly, assuming that the tasks share the same relationship across features uniformly. This assumption is seldom true as tasks may be related across some features but not others. Addressing this problem, we propose a new multi-task learning model that learns separate task relationships along different features. This added flexibility allows our model to have a finer and differential level of control in joint modeling of tasks along different features. We formulate the model as an optimization problem and provide an efficient, iterative solution. We illustrate the behavior of the proposed model using a synthetic dataset where we induce varied feature-dependent task relationships: positive relationship, negative relationship, no relationship. Using four real datasets, we evaluate the effectiveness of the proposed model for many multi-task regression and classification problems, and demonstrate its superiority over other state-of-the-art multi-task learning models

Fast one-class support vector machine for novelty detection

Novelty detection arises as an important learning task in several applications. Kernel-based approach to novelty detection has been widely used due to its theoretical rigor and elegance of geometric interpretation. However, computational complexity is a major obstacle in this approach. In this paper, leveraging on the cutting-plane framework with the well-known One-Class Support Vector Machine, we present a new solution that can scale up seamlessly with data. The first solution is exact and linear when viewed through the cutting-plane; the second employed a sampling strategy that remarkably has a constant computational complexity defined relatively to the probability of approximation accuracy. Several datasets are benchmarked to demonstrate the credibility of our framework.

Prediciton of emergency events: a multi-task multi-label learning approach

Prediction of patient outcomes is critical to plan resources in an hospital emergency department. We present a method to exploit longitudinal data from Electronic Medical Records (EMR), whilst exploiting multiple patient outcomes. We divide the EMR data into segments where each segment is a task, and all tasks are associated with multiple patient outcomes over a 3, 6 and 12 month period. We propose a model that learns a prediction function for each task-label pair, interacting through two subspaces: the first subspace is used to impose sharing across all tasks for a given label. The second subspace captures the task-specific variations and is shared across all the labels for a given task. The proposed model is formulated as an iterative optimization problems and solved using a scalable and efficient Block co-ordinate descent (BCD) method. We apply the proposed model on two hospital cohorts - Cancer and Acute Myocardial Infarction (AMI) patients collected over a two year period from a large hospital emergency department. We show that the predictive performance of our proposed models is significantly better than those of several state-of-the-art multi-task and multi-label learning methods.

Small-variance asymptotics for bayesian nonparametric models with constraints

The users often have additional knowledge when Bayesian nonparametric models (BNP) are employed, e.g. for clustering there may be prior knowledge that some of the data instances should be in the same cluster (must-link constraint) or in different clusters (cannot-link constraint), and similarly for topic modeling some words should be grouped together or separately because of an underlying semantic. This can be achieved by imposing appropriate sampling probabilities based on such constraints. However, the traditional inference technique of BNP models via Gibbs sampling is time consuming and is not scalable for large data. Variational approximations are faster but many times they do not offer good solutions. Addressing this we present a small-variance asymptotic analysis of the MAP estimates of BNP models with constraints. We derive the objective function for Dirichlet process mixture model with constraints and devise a simple and efficient K-means type algorithm. We further extend the small-variance analysis to hierarchical BNP models with constraints and devise a similar simple objective function. Experiments on synthetic and real data sets demonstrate the efficiency and effectiveness of our algorithms.

Stabilizing sparse cox model using statistic and semantic structures in electronic medical records

Stability in clinical prediction models is crucial for transferability between studies, yet has received little attention. The problem is paramount in high dimensional data, which invites sparse models with feature selection capability. We introduce an effective method to stabilize sparse Cox model of time-to-events using statistical and semantic structures inherent in Electronic Medical Records (EMR). Model estimation is stabilized using three feature graphs built from (i) Jaccard similarity among features (ii) aggregation of Jaccard similarity graph and a recently introduced semantic EMR graph (iii) Jaccard similarity among features transferred from a related cohort. Our experiments are conducted on two real world hospital datasets: a heart failure cohort and a diabetes cohort. On two stability measures – the Consistency index and signal-to-noise ratio (SNR) – the use of our proposed methods significantly increased feature stability when compared with the baselines.

Learning entry profiles of children with autism from multivariate treatment information using restricted boltzmann machines

Entry profiles can be generated before children with Autism Spectrum Disorders (ASD) begin to traverse an intervention program. They can help evaluate the progress of each child on the dedicated syllabus in addition to enabling narrowing down the best intervention course over time. However, the traits of ASD are expressed in different ways in every individual affected. The resulting spectrum nature of the disorder makes it challenging to discover profiles of children with ASD. Using data from 491 children, traversing the syllabus of a comprehensive intervention program on iPad called TOBY Playpad, we learn the entry profiles of the children based on their age, sex and performance on their first skills of the syllabus. Mixed-variate restricted Boltzmann machines allow us to integrate the heterogeneous data into one model making it a suitable technique. The data based discovery of entry profiles may assist in developing systems that can automatically suggest best suitable paths through the syllabus by clustering the children based on the characteristics they present at the beginning of the program. This may open the pathway for personalised intervention.

Nonparametric discovery of movement patterns from accelerometer signals

Monitoring daily physical activity plays an important role in disease prevention and intervention. This paper proposes an approach to monitor the body movement intensity levels from accelerometer data. We collect the data using the accelerometer in a realistic setting without any supervision. The ground-truth of activities is provided by the participants themselves using an experience sampling application running on their mobile phones. We compute a novel feature that has a strong correlation with the movement intensity. We use the hierarchical Dirichlet process (HDP) model to detect the activity levels from this feature. Consisting of Bayesian nonparametric priors over the parameters the model can infer the number of levels automatically. By demonstrating the approach on the publicly available USC-HAD dataset that includes ground-truth activity labels, we show a strong correlation between the discovered activity levels and the movement intensity of the activities. This correlation is further confirmed using our newly collected dataset. We further use the extracted patterns as features for clustering and classifying the activity sequences to improve performance.

Dynamical analysis of neural networks with time-varying delays using the LMI approach

This study is concerned with the delay-range-dependent stability analysis for neural networks with time-varying delay and Markovian jumping parameters. The time-varying delay is assumed to lie in an interval of lower and upper bounds. The Markovian jumping parameters are introduced in delayed neural networks, which are modeled in a continuous-time along with finite-state Markov chain. Moreover, the sufficient condition is derived in terms of linear matrix inequalities based on appropriate Lyapunov-Krasovskii functionals and stochastic stability theory, which guarantees the globally asymptotic stable condition in the mean square. Finally, a numerical example is provided to validate the effectiveness of the proposed conditions.

Human perception-based washout filtering using genetic algorithm

The Motion Cueing Algorithm (MCA) transforms longitudinal and rotational motions into simulator movement, aiming to regenerate high fidelity motion within the simulators physical limitations. Classical washout filters are widely used in commercial simulators because of their relative simplicity and reasonable performance. The main drawback of classical washout filters is the inappropriate empirical parameter tuning method that is based on trial-and-error, and is effected by programmers’ experience. This is the most important obstacle to exploiting the platform efficiently. Consequently, the conservative motion produces false cue motions. Lack of consideration for human perception error is another deficiency of classical washout filters and also there is difficulty in understanding the effect of classical washout filter parameters on generated motion cues. The aim of this study is to present an effortless optimization method for adjusting the classical MCA parameters, based on the Genetic Algorithm (GA) for a vehicle simulator in order to minimize human sensation error between the real and simulator driver while exploiting the platform within its physical limitations. The vestibular sensation error between the real and simulator driver as well as motion limitations have been taken into account during optimization. The proposed optimized MCA based on GA is implemented in MATLAB/Simulink. The results show the superiority of the proposed MCA as it improved the human sensation, maximized reference signal shape following and exploited the platform more efficiently within the motion constraints.

Design of distributed adaptive neural traffic signal timing controller by cuckoo search optimization

This paper focuses on designing an adaptive controller for controlling traffic signal timing. Urban traffic is an inevitable part in modern cities and traffic signal controllers are effective tools to control it. In this regard, this paper proposes a distributed neural network (NN) controller for traffic signal timing. This controller applies cuckoo search (CS) optimization methods to find the optimal parameters in design of an adaptive traffic signal timing control system. The evaluation of the performance of the designed controller is done in a multi-intersection traffic network. The developed controller shows a promising improvement in reducing travel delay time compared to traditional fixed-time control systems.

Distributed Q-learning controller for a multi-intersection traffic network

This paper proposes a Q-learning based controller for a network of multi intersections. According to the increasing amount of traffic congestion in modern cities, using an efficient control system is demanding. The proposed controller designed to adjust the green time for traffic signals by the aim of reducing the vehicles’ travel delay time in a multi-intersection network. The designed system is a distributed traffic timing control model, applies individual controller for each intersection. Each controller adjusts its own intersection’s congestion while attempt to reduce the travel delay time in whole traffic network. The results of experiments indicate the satisfied efficiency of the developed distributed Q-learning controller.

Optimal feature subset selection for neuron spike sorting using the genetic algorithm

It is crucial for a neuron spike sorting algorithm to cluster data from different neurons efficiently. In this study, the search capability of the Genetic Algorithm (GA) is exploited for identifying the optimal feature subset for neuron spike sorting with a clustering algorithm. Two important objectives of the optimization process are considered: to reduce the number of features and increase the clustering performance. Specifically, we employ a binary GA with the silhouette evaluation criterion as the fitness function for neuron spike sorting using the Super-Paramagnetic Clustering (SPC) algorithm. The clustering results of SPC with and without the GA-based feature selector are evaluated using benchmark synthetic neuron spike data sets. The outcome indicates the usefulness of the GA in identifying a smaller feature set with improved clustering performance.

GRIOT - sistema CAD inteligente para design de calçados

O presente trabalho é o estudo dos aspectos da metodologia projetual, face às novas tecnologias da informática, Inteligência Artificial (Sistemas Especialistas) e CAD, consideradas as reais possibilidades de automatização no processo de concepção em Design. Esse artigo propõe uma metodologia para a construção de sistema inteligente capaz de auxiliar o designer nas tarefas projetuais. A indústria de calçados foi utilizada como estudo de caso para a aplicação da metodologia, onde as reais possibilidades de automação são verificadas.

Color texture analysis based on fractal descriptors

Color texture classification is an important step in image segmentation and recognition. The color information is especially important in textures of natural scenes, such as leaves surfaces, terrains models, etc. In this paper, we propose a novel approach based on the fractal dimension for color texture analysis. The proposed approach investigates the complexity in R, G and B color channels to characterize a texture sample. We also propose to study all channels in combination, taking into consideration the correlations between them. Both these approaches use the volumetric version of the Bouligand-Minkowski Fractal Dimension method. The results show a advantage of the proposed method over other color texture analysis methods. (C) 2011 Elsevier Ltd. All rights reserved.