What has ‘intelligent computation’ to do with the emerging field of Human Data Interaction? This talk considers (some) of the challenges that drive the ‘decentring’ of computation from the perspective of everyday life and the need to make technical solutions accountable to ordinary people. We will consider the Databox approach and broader need to put Human Interaction centre stage alongside current investments in the Data processing ecosystem.
The communication presentations (keynote talks, facilitated break-out sessions, posters) will be continuously uploaded in May and June.
Andy Crabtree is Professor of Computer Science at the University of Nottingham. He is a social scientist by background, specialising in ethnography for systems design. He is an EPSRC Fellow and his current research focuses on making the Internet of Things accountable to ordinary users and developing the Databox platform to enable this.
Dr. Hava Siegelmann joined DARPA in July 2016 with the goal of developing programs that advance intelligence in computerized devices, focusing on life-long learning, context-aware adaptivity, and user-centered applications. Prior to joining DARPA, Dr. Siegelmann directed the Biologically Inspired Neural and Dynamical Systems (BINDS) Laboratory at the University of Massachusetts Amherst, from which she is on leave. While at the University, she also served as a Core Member of the Neuroscience and Behavior Program. Dr. Siegelmann’s mathematical and computational studies of the brain, somatic cells, cognition, and intelligence depend on a multi-disciplinary approach that combines complexity science, information and learning theories, computational simulations, biology, and neural networks. A unifying theme of her work has been the study of time-dependent adaptive dynamical complex systems. One of her research goals involves further investigation of how an underlying architecture brings about the dynamics that evolve into intelligent behavior and how behavioral feedback from the dynamics proceeds toward adaptation in the architecture. Her research accomplishments include advancing the understanding of biologically-inspired computational systems, among them neural systems and genetic networks of organisms. Dr. Siegelmann co-originated Support Vector Clustering, which has become one of the most widely used clustering algorithms in industry. She also created a sub-field of computation with her discovery of Super-Turing computation theory, which continues to spawn innovations in both computational methods and the interpretation of cognitive, biological, and physical processes.
The exponential growth following Moore's law will come to an end with transistors. Other options include biotechnology, nanotechnology and analog computation. Analog computation provides very low-power and cheap solution to the simulation of neural networks. It offers higher levels of expressivity than Turing Machines and is required for lifelong learning machines.
Jerzy Gorecki is a professor at the Institute of Physical Chemistry, Polish Academy of Sciences, Warsaw. His research areas are:
- Unconventional computing with nonequilibrium chemical medium.
- Computational chemical kinetics; large scale microscopic computer simulations of chemical systems far-from-equilibrium.
- Stochastic effects in chemical systems.
J. Gorecki, K. Gizynski and L. Zommer
The top-down approach is applied to construct database classifiers performing information processing with a network of coupled chemical oscillators. Oscillators are described by a model of a photosensitive variant of Belousov-Zhabotinsky reaction and their activity is controlled by an external illumination. The input information is introduced by illumination of selected oscillators. The output information is extracted from the number of oscillations observed at the chosen network element. An evolutionary algorithm is used to determine the control (illumination time) of other droplets in the network to maximize the mutual information between the types of database records and the network outputs.
chemical oscillations, database classification, top-down design, evolutionary algorithm, network
Dr. José Santa received in (1999-)2004 the M.S. degree (5-year) in Computer Engineering at the University of Murcia, Spain; in 2008 he received the M.S. degree in Advanced Information and Telematics Technologies; in 2009 he obtained his PhD in Computer Science at University of Murcia. A great part of his research work, both before and after his PhD, is about intelligent transportation systems (ITS), mobile communications, next-generation networks and cyber-physical systems, with special emphasis on real prototypes and evaluation. He has participated and coordinated international and national projects, such as the EU GIROADS, ITSSv6, FOTsis and the Spanish OASIS, TIMI, m:Via and S-CICLO, among others. Moreover, he has been part of the COST actions ARTS and WISE-ACT, in the last one as member of the management committee. He is a regular participant on international conferences, and he has published works in high impact journals about his research. He is a regular member of relevant international committees in conferences and journals in the area. Currently he is a Senior Research Fellow at Department of Information and Communications Engineering, at University of Murcia.
Data processing can be delegated nowadays among potential fog nodes, edge servers, cloudlets or cloud datacentres, involving computing paradigms that have appeared to deal with data generation volumes and software needs of constrained and mobile final devices in smart and Internet of Things (IoT) scenarios. However, the future of computing appears not to fit well in a three-tier discrete distribution, but a whole path of computing nodes at different levels. In this line, there is a need of novel intelligent offloading and coordination strategies in the new arena of computing abstractions that identify network nodes where to perform processing tasks in a flexible way.
Following this rationale, this talk explores the current status of data gathering network architectures and present the most prominent challenges in the area to deal with computing spreading along the network path. Useful technologies and research disciplines to cover such challenging research topic are discussed, and some initial steps covered in this direction are presented.
Josep Puyol-Gruart is a researcher of the Departament de Sistemes Multiagent at the Artificial Intelligence Research Institute (IIIA).
Josep Puyol-Gruart, Carles Sierra
Artificial agents are entities with some kind of body and a set of sensors and actuators to perceive and modify the environment. The world is inherently analog and then has to be the signals received by the sensors and the interaction produced by the actuators. We introduce the Artificial Intelligence Research Institute (IIIA) and the main research that can be benefited from analog computation.
Multiagent systems, machine learning, approximate reasoning, knowledge-based systems, applications.
Nazmiye Balta-Ozkan is Senior Lecturer in Environmental/Energy Economics at Centre for Environment and Agricultural Informatics, Cranfield University, UK. Her main research interests include energy systems analysis; integration of environment, economy and energy models; social construction of smart grids across space and time as well as its implications for policy. With a background in urban and regional planning, she is interested in understanding the interactions between social, economic, environmental and technological systems, from household up to network and city level.
Nazmiye Balta-Ozkan, Simon Tindemans, David Corne, Goran Strbac, Lorraine Whitmarsh, Lynne Baillie, Mike Just, Mike Tyler
The roll-out of smart meters and the emergence of internet of things, along with the increasing availability of new forms of user data from crowdsourced platforms such as social media, mobile phones and apps offer an immense opportunity to understand consumers’ energy behaviours and preferences and changes in energy mix in near real-time. Yet, increasing connectivity across transport, communications and energy systems mean insights from energy research on demands for energy can improve our understanding of spatial organisation of economic activities. This research identifies some of the opportunities and challenges presented by digitisation of energy systems.
smart grid, big data, networked infrastructure, spatial economy
Nicholas Race is Professor of Networked Systems at Lancaster University. His research focuses on developing future networking services built upon Software Defined Networks and Network Functions Virtualisation. This includes new techniques to enhance the Quality of Experience of media streaming and support for the detection and remediation of network anomalies. He leads NG-CDI - a new EPSRC/BT funded project that aims to develop a future network that is “autonomic”, with the capability to react and reconfigure infrastructure accordingly with minimal human intervention.
Traditionally the deployment of new networking services has involved reinvestment in infrastructure, extensive pre-testing, and people-intensive service support in operation. Future services will change ever more rapidly – and unpredictably – and therefore a fundamental change is needed to the way networking infrastructure and associated services are developed. Software Defined Networking (SDN) and Network Functions Virtualisation (NFV) are key approaches to improving the agility and responsiveness of network infrastructure. The talk introduces the potential of SDN/NFV as enabling technologies and explains how an Orchestrator will be key to the successful rollout of Virtual Network Functions (VNFs). The exciting opportunities offered by Fog/Edge computing go beyond current thinking in the rollout of VNFs and orchestrators – therefore the presentation will highlight a number of areas where further research is required.
Sayani Majumdar is an academy research fellow of the Departmentof Applied Physics at Aalto University. Her research is to clarify the spin physics in FM half-metals, ferroelectric and organic semiconductors to enable their application in hybrid spintronic components. The inorganic-organic interface in the hybrid structures are of utmost importance. Understanding of the basic spin-physics and spin charge correlation in these materials will lead to better clarification of the spin-dynamics paving their way to cost-effective, multi-functional, energy-efficient electronics in future.
Sayani Majumdar, Hongwei Tan, Qihang Qin and Sebastiaan van Dijken
Neuromorphic computing is the state-of-the-art research trend in the field of memory and logic devices where the goal is to build a versatile computer that is efficient in terms of energy and space, homogeneously scalable to large networks of neurons and synapses, and flexible enough to run complex behavioral models of the neocortex as well as networks inspired by neural architectures. Memristors, with their gradually changing conductivity can mimic the biological synapses. Low energy consumption, ultrafast operation, large number of conducting states and small dimensions are the most essential requirements for a memristor to perform tasks like a synapse. A ferroelectric tunnel junction (FTJ), where gradual modulation of conductance can be achieved by controlled rotation of ferroelectric domains, can act efficiently as a synapse. Here, we report synaptic functions can be emulated efficiently in FTJs consisting of organic ferroelectric P(VDF-TrFE) tunnel barrier. Large on/off ratio at room temperature together with reproducible memristive behavior, fast switching, long data retention of intermediate conductive states and spike-time-dependent plasticity based Hebbian and ani-Hebbian learning makes them extremely promising for neuromorphic applications.
Analog computing, neuromorphic devices, synaptic computation, memristors, ferroelectric tunnel junctions.
- 1 of 2
- next ›