Data augmentation for Human Activity Recognition with Generative Adversarial Networks.

Currently, Human Activity Recognition (HAR) applications need a large volume of data to be able to generalize to new users and environments. However, the availability of labeled data is usually limited and the process of recording new data is costly and time-consuming. Synthetically increasing datasets using Generative Adversarial Networks (GANs) has been proposed, outperforming cropping, time-warping, and jittering techniques on raw signals. Incorporating GAN-generated synthetic data into datasets has been demonstrated to improve the accuracy of trained models. Regardless, currently, there is no optimal GAN architecture to generate accelerometry signals, neither a proper evaluation methodology to assess signal quality or accuracy using synthetic data. This work is the first to propose conditional Wasserstein Generative Adversarial Networks (cWGANs) to generate synthetic HAR accelerometry signals. Furthermore, we calculate quality metrics from the literature and study the impact of synthetic data on a large HAR dataset involving 395 users. Results show that i) cWGAN outperforms original Conditional Generative Adversarial Networks (cGANs), being 1D convolutional layers appropriate for generating accelerometry signals, ii) the performance improvement incorporating synthetic data is more significant as the dataset size is smaller, and iii) the quantity of synthetic data required is inversely proportional to the quantity of real data.

Improving drug discovery through parallelism.

Compound identification in ligand-based virtual screening is limited by two key issues: the quality and the time needed to obtain predictions. In this sense, we designed OptiPharm, an algorithm that obtained excellent results in improving the sequential methods in the literature. In this work, we go a step further and propose its parallelization. Specifically, we propose a two-layer parallelization. Firstly, an automation of the molecule distribution process between the available nodes in a cluster, and secondly, a parallelization of the internal methods (initialization, reproduction, selection and optimization). This new software, called pOptiPharm, aims to improve the quality of predictions and reduce experimentation time. As the results show, the performance of the proposed methods is good. It can find better solutions than the sequential OptiPharm, all while reducing its computation time almost proportionally to the number of processing units considered.

Black-box and surrogate optimization for tuning spiking neural models of striatum plasticity.

The basal ganglia (BG) is a brain structure that has long been proposed to play an essential role in action selection, and theoretical models of spiking neurons have tried to explain how the BG solves this problem. A recently proposed functional and biologically inspired network model of the striatum (an important nucleus of the BG) is based on spike-timing-dependent eligibility (STDE) and captured important experimental features of this nucleus. The model can recognize complex input patterns and consistently choose rewarded actions to respond to such sensory inputs. However, model tuning is challenging due to two main reasons. The first is the expert knowledge required, resulting in tedious and potentially biased trial-and-error procedures. The second is the computational cost of assessing model configurations (approximately 1.78 h per evaluation). This study addresses the model tuning problem through numerical optimization. Considering the cost of assessing solutions, the selected methods stand out due to their low requirements for solution evaluations and compatibility with high-performance computing. They are the SurrogateOpt solver of Matlab and the RBFOpt library, both based on radial basis function approximations, and DIRECT-GL, an enhanced version of the widespread black-box optimizer DIRECT. Besides, a parallel random search serves as a baseline reference of the outcome of opting for sophisticated methods. SurrogateOpt turns out to be the best option for tuning this kind of model. It outperforms, on average, the quality of the configuration found by an expert and works significantly faster and autonomously. RBFOpt and the random search share the second position, but their average results are below the option found by hand. Finally, DIRECT-GL follows this line becoming the worst-performing method.

A two-layer mono-objective algorithm based on guided optimization to reduce the computational cost in virtual screening.

Virtual screening methods focus on searching molecules with similar properties to a given compound. Molecule databases are made up of large numbers of compounds and are constantly increasing. Therefore, fast and efficient methodologies and tools have to be designed to explore them quickly. In this context, ligand-based virtual screening methods are a well-known and helpful tool. These methods focus on searching for the most similar molecules in a database to a reference one. In this work, we propose a new tool called 2L-GO-Pharm, which requires less computational effort than OptiPharm, an efficient and robust piece of software recently proposed in the literature. The new-implemented tool maintains or improves the quality of the solutions found by OptiPharm, and achieves it by considerably reducing the number of evaluations needed. Some of the strengths that help 2L-GO-Pharm enhance searchability are the reduction of the search space dimension and the introduction of some circular limits for the angular variables. Furthermore, to ensure a trade-off between exploration and exploitation of the search space, it implements a two-layer strategy and a guided search procedure combined with a convergence test on the rotation axis. The performance of 2L-GO-Pharm has been tested by considering two different descriptors, i.e. shape similarity and electrostatic potential. The results show that it saves up to 87.5 million evaluations per query molecule.

DOLARS, a Distributed On-Line Activity Recognition System by Means of Heterogeneous Sensors in Real-Life Deployments-A Case Study in the Smart Lab of The University of Almería.

Activity Recognition (AR) is an active research topic focused on detecting human actions and behaviours in smart environments. In this work, we present the on-line activity recognition platform DOLARS (Distributed On-line Activity Recognition System) where data from heterogeneous sensors are evaluated in real time, including binary, wearable and location sensors. Different descriptors and metrics from the heterogeneous sensor data are integrated in a common feature vector whose extraction is developed by a sliding window approach under real-time conditions. DOLARS provides a distributed architecture where: (i) stages for processing data in AR are deployed in distributed nodes, (ii) temporal cache modules compute metrics which aggregate sensor data for computing feature vectors in an efficient way; (iii) publish-subscribe models are integrated both to spread data from sensors and orchestrate the nodes (communication and replication) for computing AR and (iv) machine learning algorithms are used to classify and recognize the activities. A successful case study of daily activities recognition developed in the Smart Lab of The University of Almería (UAL) is presented in this paper. Results present an encouraging performance in recognition of sequences of activities and show the need for distributed architectures to achieve real time recognition.

Solving the multiple competitive facilities location and design problem on the plane.

A continuous location problem in which a firm wants to set up two or more new facilities in a competitive environment is considered. Other facilities offering the same product or service already exist in the area. Both the locations and the qualities of the new facilities are to be found so as to maximize the profit obtained by the firm. This is a global optimization problem, with many local optima. In this paper we analyze several approaches to solve it, namely, three multistart local search heuristics, a multistart simulated annealing algorithm, and two variants of an evolutionary algorithm. Through a comprehensive computational study it is shown that the evolutionary algorithms are the heuristics that provide the best solutions. Furthermore, using a set of problems for which the optimal solutions are known, only the evolutionary algorithms were able to find the optimal solutions for all the instances. The evolutionary strategies presented in this paper can be easily adapted to handle other continuous location problems.