Much recent research has been conducted in the area of Bayesian learning, particularly with regard to the optimization of hyper-parameters via Gaussian process regression. The methodologies rely chiefly on the method of maximizing the expected improvement of a score function with respect to adjustments in the hyper-parameters. In this work, we present a novel algorithm that exploits notions of confidence intervals and uncertainties to enable the discovery of the best optimal within a targeted region of the parameter space. We demonstrate the efficacy of our algorithm with respect to machine learning problems and show cases where our algorithm is competitive with the method of maximizing expected improvement.

Nonlinear systems of differential equations have attracted the interest in fields like system biology, ecology or biochemistry, due to their flexibility and their ability to describe dynamical systems. Despite the importance of such models in many branches of science they have not been the focus of systematic statistical analysis until recently. In this work we propose a general approach to estimate the parameters of systems of differential equations measured with noise. Our methodology is based on the maximization of the penalized likelihood where the system of differential equations is used as a penalty. To do so, we use a Reproducing Kernel Hilbert Space approach that allows us to formulate the estimation problem as an unconstrained numeric maximization problem easy to solve. The proposed method is tested with synthetically simulated data and it is used to estimate the unobserved transcription factor CdaR in Steptomyes coelicolor using gene expression data of the genes it regulates. Keywords: System of ordinary differential equations, differential operator, reproducing kernel Hilbert space, gene regulatory network 1. Introduction Despite the fact that differential equations are a common modelling tool within science and engineering, statistical methods for estimating such models have only received widespread attention during the last few years. The difficulty of solving differential equations in general has been a major stumbling block for efficient statistical procedures.

Dialogue games are a two-player semantics for a variety of logics, including intuitionistic and classical logic. Dialogues can be viewed as a kind of analytic calculus not unlike tableaux. Can dialogue games be an effective foundation for proof search in intuitionistic logic (both first-order and propositional)? We announce Kuno, an automated theorem prover for intuitionistic first-order logic based on dialogue games.

A highly comparative, feature-based approach to time series classification is introduced that uses an extensive database of algorithms to extract thousands of interpretable features from time series. These features are derived from across the scientific time-series analysis literature, and include summaries of time series in terms of their correlation structure, distribution, entropy, stationarity, scaling properties, and fits to a range of time-series models. After computing thousands of features for each time series in a training set, those that are most informative of the class structure are selected using greedy forward feature selection with a linear classifier. The resulting feature-based classifiers automatically learn the differences between classes using a reduced number of time-series properties, and circumvent the need to calculate distances between time series. Representing time series in this way results in orders of magnitude of dimensionality reduction, allowing the method to perform well on very large datasets containing long time series or time series of different lengths. For many of the datasets studied, classification performance exceeded that of conventional instance-based classifiers, including one nearest neighbor classifiers using Euclidean distances and dynamic time warping and, most importantly, the features selected provide an understanding of the properties of the dataset, insight that can guide further scientific investigation.

User profiles have an important role in multi-agent recommender systems. The information stored in them improves the system's generated recommendations. Multi-agent recommender systems learn from previous recommendations to update users' profiles and improving next recommendations according to the user feedback. However, when the user does not evaluate the recommendations the system may deliver poor recommendations in the future. This paper presents a mechanism that explores user information from social networks to update the user profile and to generate implicit evaluations on behalf of the user. The mechanism was validated with travel packages recommendations and some preliminary results illustrate how user information gathered from social networks may help to improve recommendations in multi-agent recommender systems.

The Semantic Sensor Web (SSW) will eventually revolutionize how we perceive and query information about the physical world. Currently, there is an ongoing effort to develop a searchable web of things that sense and control the world. As this new internet of things expands, there will be an explosion of available raw data that may not always be reachable by users. Bridging this gap between what the user wants and the information collected and represented by embedded devices is a critical issue. In order to really maximize the benefit of such a web of networked sensing devices, new semantic approaches that can infer and predict additional information about the sensors and their context need to be developed. This paper proposes a content enrichment approach that uses sensor and context data as features to predict new meta-tags that can further identify relevant categorizations for the embedded devices and the physical data they collect. Specifically, machine learning classification and regression techniques are used to predict semantic tags for each embedded system context. Results of the 10-fold cross validation analysis and feature ranking are presented and discussed.

This work investigates how wireless signal attenuation affects a team of mobile robots performing exploration. Many coverage tasks, such as search and rescue or exploration, can be performed more effectively when robots communicate with one other. However, in real world environments, maintaining communication can be difficult due to unpredictable wireless signal propagation. In this paper, we investigate the wireless network connectivity in an outdoor area that consist of concrete wall and pillars. Preliminary simulation and physical experimental results are compared to demonstrate the effects of signal attenuation on robot team performance.

There is a growing interest within educational research to produce culturally-aware intelligent learning environments (ILEs) that capitalize on the affective benefits of positive cultural resonance and avoid the counter-productive effects of culturally ignorant designs. Several challenges arise when attempting to produce culturally-appropriate content for ILEs. These stem from the need for semantic representations of cultural conceptualisations that go beyond folk approaches, have sufficient details for intracultural reasoning, and which can be matched with the cultural backgrounds of students who use these ILEs. This paper tackles these challenges firstly through the formalism of a lower-level ontology for describing the cultural semantics commonly used in educational content and secondly with a software component for reasoning about this ontological knowledge in relation to student cultural backgrounds. An application was developed to test the practicality of the approach and assess its utility in locating culturally-appropriate educational resources for students. The evaluation results revealed that the majority of content selections made by the system were rated as highly appropriate by 90% of the participants on average and confirmed the viability of the approach.

High dimensionality (having a large number of independent attributes) is a major problem for bioinformatics datasets such as gene microarray datasets. Feature selection algorithms are necessary to remove the irrelevant (not useful) and redundant (contain duplicate information) features. One approach to handle this problem is wrapper-based subset evaluation, which builds classification models on different feature subsets to discover which performs best. Although the computational complexity of this technique has led to it being rarely used for bioinformatics, its ability to find the features which give the best model make it important in this domain. However, when using wrapper-based feature selection, it is not obvious whether the learner used within the wrapper should match the learner used for building the final classification model. Furthermore, this question may depend on other properties of the dataset, such as difficulty of learning (general performance without feature selection) and dataset balance (ratio of minority and majority instances). To study this, we use nine datasets with varying levels of difficulty and balance. We find that across all datasets, the best strategy is to use one learner (Na¨ıve Bayes) inside the wrapper regardless of the learner which will be used outside. However, when broken down by difficulty and balance levels, our results show that the more balanced and less difficult datasets work best when the learners inside and outside the wrapper match. Thus, the answer to this question will depend on properties of the dataset.

With the recent popularity of electronic medical records, enormous amount of medical data is being generated every day at an exponential rate.Machine learning methods have been shown in many studies to be capable of producing automatic medical diagnostic models such as automated prognostic models. However, many powerful machine learning algorithms such as support vector machine (SVM), Random Forest (RF) or Kernel Logistic Regression (KLR) are unbearably slow for very large datasets. This makes their use in medical research limited to small to medium scale problems.This study is motivated by an ongoing research on prostate cancer mortality prediction for a national representative of US population where the SVM and RF took several hours or days to trainwhereas simple linear methods such as logistic regression or linear discriminant analysis take minutes or even seconds.Because, most real-world problems are non-linear, this paper presents a large scale algorithm enabling a recently proposed least squares extreme logistic regression to learn very large datasets. The algorithm is shown on a case study of mortality prediction for men diagnosed with early stage prostate cancer to provide very fast and more accurate result than standard statistical methods.