Intrinsic dynamics within the brain can accelerate learning by providing a prior scaffolding for dynamics aligned with task objectives. Such intrinsic dynamics would ideally self-organize and self-sustain in the face of biological noise including synaptic turnover and cell death. An example of such dynamics is the formation of sequences, a ubiquitous motif in neural activity. The sequence-generating circuit in zebra finch HVC provides a reliable timing scaffold for motor output in song and demonstrates a remarkable capacity for unsupervised recovery following perturbation. Inspired by HVC, we seek a local plasticity rule capable of organizing and maintaining sequence-generating dynamics despite continual network perturbations. We adopt a meta-learning approach introduced by Confavreux et al, which parameterizes a learning rule using basis functions constructed from pre- and postsynaptic activity and synapse size, with tunable time constants. Candidate rules are simulated within initially random networks, and their fitness is evaluated according to a loss function that measures the fidelity with which the resulting dynamics encode time. We use this approach to introduce biological noise, forcing meta-learning to find robust solutions. We first show that, in the absence of perturbations, meta-learning identifies a temporally asymmetric generalization of Oja's rule that reliably organizes sparse sequential activity. When synaptic turnover is introduced, the learned rule incorporates a form of homeostasis, better maintaining robust sequential dynamics relative to other previously proposed rules. Additionally, inspired by recent findings demonstrating that the strength of projections from inhibitory interneurons in HVC also dynamically responds to perturbations, we explore the role of inhibitory plasticity in sequence-generating circuits. We find that learned plasticity adjusts both excitation and inhibition in response to manipulations, outperforming rules applied only to excitatory connections. We demonstrate how plasticity acting on both excitatory and inhibitory synapses can better shape excitatory cell dynamics to scaffold timing representations.

The quantum computing community has been searching for suitable applications to demonstrate the potential of near-term quantum devices. Quantum machine learning is a potential candidate, particularly using models that cannot be efficiently simulated with classical computers [1] , [2] . This work focuses on a transition phase of quantum computers where the quantum machine learning model is still simulable classically but projected not to be simulable as the size of the model grows. Ultimately quantum computers may have advantages for high-dimensional real-world problems. Due to the limited number of qubits in current noisy intermediate-scale quantum (NISQ) devices, the direct application of quantum computers in high dimensional data is not feasible. To remedy this problem, an encoder-decoder architecture can be utilized. The encoder model would transform the high-dimensional data into a compact representation, to a level that small quantum computers can be used today (or in the near future), and the decoder would take the quantum processed outputs back to the high-dimensional space [3] .

In this paper, a novel particle swarm optimization (PSO) algorithm is proposed in order to improve the accuracy of traditional clustering approaches with applications in analyzing real-time patient attendance data from an accident & emergency (A&E) department in a local U.K. hospital. In the proposed randomly occurring distributedly delayed PSO (RODDPSO) algorithm, the evolutionary state is determined by evaluating the evolutionary factor in each iteration, based on whether the velocity updating model switches from one mode to another. With the purpose of reducing the possibility of getting trapped in the local optima and also expanding the search space, randomly occurring time-delays that reflect the history of previous personal best and global best particles are introduced in the velocity updating model in a distributed manner. Eight well-known benchmark functions are employed to evaluate the proposed RODDPSO algorithm which is shown via extensive comparisons to outperform some currently popular PSO algorithms. To further illustrate the application potential, the RODDPSO algorithm is successfully exploited in the patient clustering problem for data analysis with respect to a local A&E department in West London. Experiment results demonstrate that the RODDPSO-based clustering method is superior over two other well-known clustering algorithms.

In this paper we describe how to use a Game-Based Learning (GBL) approach to help end-user programmers to develop Computational Thinking (CT) skills. CT represents a framework consisting of a set of skills such as: abstraction, decomposition and more in general problem solving expressed in a computational form (algorithmic thinking) that can be easily formalised into a computer program. Those skills are beneficial for end-user programmers in order to model a task for being tackled with computational tools, e.g. spreadsheets, domain specific and visual languages. TAPASPlay is an engaging GBL platform designed to help end-user to learn algorithmic thinking.

Customer retention is of crucial importance for companies in the mobile services industry (MSI) because of the increasing competition in the mobile services sector. As a result, companies are increasingly exploiting data to address customer retention. Traditional data collection methods such as questionnaires and interviews have been utilised to address customer retention. In addition, data mining techniques such as classification and clustering have been applied to understand customer retention. The effectiveness of utilising these techniques to investigating customer retention is debatable as a result of the complexity of the mobile services market. This study proposes a novel data-driven methodological approach to investigating customer retention in the MSI, using agent based modelling and simulation (ABMS). The dataset for this study is extracted from Twitter using specific keywords to gather data from mobile services companies of interest. The design science paradigm is adopted as the overarching methodology along with a selected framework, the social media domain analysis framework (SoMeDoA) to generate customer satisfaction determinants and agent models. ABMS is performed with the established determinants to uncover the social effects of customer retention. Data analysis show that customers' environment is a key influence on their decision to remain with (or leave) their mobile services provider, with word of mouth as an important factor. Importantly, ABMS is able to explore further determinants in a wider market place.

This paper presents a novel approach to social data analysis, exploring the use of microblogging to manage interaction between humans and robots, and presenting and evaluating an architecture that extends the use of social networks to connect humans and devices. The approach uses natural language processing – in the form of simple grammar-based techniques – to extract features of interest from textual data retrieved from a microblogging platform in real-time and generate appropriate executable code for the robot. The simple rule-based solution exploits some of the ‘natural’ constraints imposed by microblogging platforms to manage the potential complexity of the interactions and create bi-directional communication. In order to evaluate the developed system, an analysis of real-time, user-generated social media data is presented. The analysis demonstrates the feasibility of producing programmes from the social media data which lead to executable actions by a front-end application – an approach of immediate relevance to web-based systems, like question–answering engines, personal digital assistants, and smart home/office devices.

COTS simulation packages (CSPs) have proved popular in an industrial setting with a number of software vendors. In contrast, options for re-using existing models seem more limited. Re-use of simulation component models by collaborating organizations is restricted by the same semantic issues however that restrict the inter-organization use of web services. The current representations of web components are predominantly syntactic in nature lacking the fundamental semantic underpinning required to support discovery on the emerging semantic web. Semantic models, in the form of ontology, utilized by web service discovery and deployment architecture provide one approach to support simulation model reuse. Semantic interoperation is achieved through the use of simulation component ontology to identify required components at varying levels of granularity (including both abstract and specialized components). Selected simulation components are loaded into a CSP, modified according to the requirements of the new model and executed. The paper presents the development of ontology, connector software and web service discovery architecture in order to understand how such ontology are created, maintained and subsequently used for simulation model reuse. The ontology is extracted from health service simulation - comprising hospitals and the National Blood Service. The ontology engineering framework and discovery architecture provide a novel approach to inter- organization simulation, uncovering domain semantics and adopting a less intrusive interface between participants. Although specific to CSPs the work has wider implications for the simulation community.