Monday, 29 July 2019
How can artificial life help solve societal challenges? A part of the answer lies into understanding human behavior at the root of these challenges and society’s responses when faced to them. A large body of work investigates the causes and solutions of such challenges but their results often remain in the realm of theory. This workshop proposes to explore how research on the evolution of human behavior can be moved beyond theoretical realm to address societal challenges.
The workshop will be organised in two parts. The first 1.5 hours will focus on the state-of-the-art theoretical and modelling work. The second 1.5 hours will discuss how this knowledge on the evolution of human behavior can contribute to (i) improve human societies, (ii) design artificial societies and (iii) develop the interface between them.
The workshop will emphasize the role of new technology which can either conflict with previous behavioural adaptations or shape our cultural evolution. It includes topics such as:
Tuesday, 30 July 2019
The fields of social learning and cultural evolution aim at understanding how the exchange of knowledge within a group of individuals influences their performance. While cultural evolution focuses on how collective knowledge evolves over time within a population, social learning is concerned with the exchange of knowledge among individuals. Cultural evolution builds upon the mechanisms offered by social learning. This relationship is visible within the Alife community as social learning and cultural evolution are studied using similar methodologies, such as evolutionary robotics, evolutionary game theory, and evolutionary algorithms. This edition of the workshop will mainly, but not exclusively, focus on grand challenges of social learning and cultural evolution research in artificial life.
Friday, 2 August 2019
The recent explosion of online big data has allowed us to conduct high-resolution quantitative modeling and analysis of complex human social behaviors. This has completely changed the way of studying collective behaviors of humans in social systems. Conventional approaches that attempted to model human behaviors using mathematically rigorous yet idealized game theoretic frameworks often failed to predict real social dynamics, because humans are often irrational and do not follow well-defined decision making protocols. Big data analytics provides us with more empirical methods and tools to capture and analyze such realistic human behaviors.
Meanwhile, this change does not necessarily mean the traditional theoretical frameworks such as game theory have lost values in scientific research on human behaviors. What is more important than ever, is to merge and integrate these multiple approaches to develop a comprehensive methodology for the understanding of human social behaviors. The key ingredient that transcends methodological variations is “computation.”
The aim of this workshop is to provide a transdisciplinary venue where researchers within and outside the Artificial Life community can come together, share their cutting-edge research on data-driven approaches and/or theoretical/mathematical approaches to human social behaviors, and collectively develop an integrated, comprehensive computational methodology for the understanding of real-world human social dynamics. Of particular interest at this workshop, as an example of hybrids of multiple approaches, is in the hybrid coordination of autonomous software agents and real humans, which has a potential to show better performance in practical problem solving than only humans or software alone. Other computational and/or hybrid approaches will also be explored.
Thursday, 1 August 2019
This workshop aims to bring together researchers who are interested in using agent-based modelling to understand human behaviour. It is a combination of agent-based modelling and behavioural science, which is a new and growing area of research. We want to build a focused group of people, bringing together many of the researchers in this young field for the first time. Agent-based modelling has a long history of success in many related fields from economics and cooperative behaviours, to social conflict, civil violence and revolution. However, its use remains very limited in studies of how human interaction is affected by more complex aspects of human behaviour, such as personality, emotional state and conflict.
In the fields of Alife, Evolutionary Computing and Agent-Based Simulation, researchers have modelled various aspects of ecosystems such as evolutionary dynamics within interacting populations. Classic works in this area include studies by Axelrod and Hamilton on the evolution of cooperation and Maynard Smith and Price on conflicts between animals of the same species. More recently, Holland created Echo, a generic ecosystem model in which evolving agents are situated in a resource-limited environment.
The motivation behind this work is to address some of the significant issues in psychological research today. Human experimentation can create ethical issues and has been increasingly difficult to conduct, making it more difficult to progress our understanding in the area. Alife models offer the capability to create realistic laboratories for which to conduct such experiments. A workshop in this area can encourage Artificial Life practitioners to use behavioural modelling to assess, challenge or even replace competing theories of human behaviour. We anticipate that by introducing rigorous computational modelling to this contentious area, it will help strengthen the field of psychology and related human behavioural sciences.
Monday, 29 July 2019
Hypothesis of pre-biotic self-organization and chemical evolution leading to the emergence of life are often described in the framework of chemical reaction networks. Recently, it has been proposed that reaction networks, and other frameworks based on processes such as molecular logics, petri nets, and others, can be exploited to represent self-organizing phenomena in other domains such as ecological, cognitive and social systems. Namely, process modeling frameworks are specially suited for representing situations where different types of entities interact in contextual ways leading to the emergence of meta-structures. At an abstract level, a process framework represents a universe whose evolution corresponds to the transformation of collections of entities into other collections of entities. Hence, the collection of processes that can self-maintain themselves long enough to be observed corresponding to the self-organizing structures that emerge from the basic processes. We propose a workshop to discuss why and how these frameworks can be used to represent different kinds of interactions and the emergence of structures from such basic interactions in different domains. On the one hand, we will present lectures of foundational kind, i.e. explaining the basics of these frameworks as well as some relevant mathematical and algorithmic issues, and on the other hand we will present novel applications related to areas such as ecology, evolution, sustainability, and others.
Monday, 29 July 2019
In nature, brains are built through a process of biological development in which many aspects of the network of neurons and connections change and are shaped by external information received through sensory organs. Biological mechanisms such as axon guidance and dendrite pruning have been shown to rely on neural activity, and correlations between learning and neural topology changes, such as the growth of grey matter, have been made. Despite this, most artificial neural network (ANN) models do not include developmental mechanisms and regard learning as the adjustment of connection weights, while some that do use development restrain it to a period before the ANN is used. As ANNs have recently become a central component to both machine learning, with the success of deep learning , and bio-inspired AI, in the form of HyperNEAT, it is worthwhile to understand the cognitive functions offered by development currently missing in ANNs.
Incorporating development into ANNs raises fundamental questions. How should models of developmental neurons be represented? How frequently should development take place over the use of a neural network that is actively learning? Some of these questions have been explored in literature, such as with Cellular Automata. A special session at ICANN 2008 on constructive neural networks discussed many such questions. More recently, progressive neural networks have been used to approach multi-task learning with deep neural networks. However, many of these explorations precede the advent of modern neural networks, and modern approaches do not sufficiently answer the questions surrounding development.
Monday, 29 July 2019
The workshop “CHEMALIFORMS: Chemistry and artificial life forms” will focus on the life-like forms created in laboratory using chemical and biochemical materials. The speakers will not only present the state-of-the-art of wet artificial life, but the questions related to future challenges such as “How can Wet Artificial Life help to solve Societal Challenges” will be discussed.
Tuesday, 30 July 2019
Much of the focus within synthetic biology has been at the molecular and cellular levels, in which the bounds are drawn around the autonomous cell. Similarly, the majority of work within the artificial life community on living technologies and proto-living systems has focussed on the cell or proto-cell, with limited attention to the ecology and inevitable system-environment interactions. These approaches often struggle to produce stable and resilient systems and miss the opportunity to achieve this by leveraging the richness of ecological interactions associated with living systems and the emergent possibilities. Within the artificial life community, however, there is a long history of considering living and proto systems in all their interactions, with philosophical and engineering approaches most suited to emergent and evolving systems whose boundaries and function may not be well-defined. We believe that the tools and perspectives of this sort of approach have much more to offer towards the practice of “synthetic ecology” than more traditional reductive approaches, which try to “engineer out” complexity.
The aim of this workshop is to bring together an exciting cross section of people from the fields of synthetic ecology, microbial ecology, artificial life and beyond to develop and consolidate emerging new approaches and ways of thinking to design and manipulate microbial ecosystems. Our scope spans naturally-occuring communities to synthetic ecosystems, as well as artificially-selected or manipulated “hybrid” systems.
Tuesday 30 July 2019
Since its origins around 1990 (unofficially even earlier), the Artificial Life scientific community has investigated how to extract the “logical form of living systems” (Christopher Langton, First ALife Conference, 1989) by using soft, hard, and wet systems as support for the investigations. Since this is such a broad topic, ALife researchers focus on a range of specific questions within it. In many cases, the common thread uniting their work is the use of similar scientific methodologies and tools, based on both experimental and analytical approaches. Indeed,besides contributing to our knowledge on the logical form of living systems, ALife research has produced a number of powerful research tools and methodologies. This includes experimental methodologies in Artificial Evolution, Swarm Robotics, Swarm Chemistry, Social Learning, statistical tools for ALife experimental data and modelling techniques for ALife systems among others. The goal of the first Workshop on the Methodology in Artificial Life (MethAL Workshop) is to bring together people who have designed innovative methods for studying Artificial Life in a context where discussion can focus on these methods specifically, rather than the results they produced.
This workshop is intended to be a platform to present and discuss current approaches, advances in implementation, future vision for methodologies, and reflections of past implementations. We will aim to provide an opportunity to meet people with interests in formalizing approaches, to be exposed to current research methodologies (with emphasis on hybridization of modelling techniques) and to exchange ideas in an informal setting. The proposed structure is as follows: Opening remarks, Keynote Speaker, Contributed and extended abstracts, Discussions and end remarks. Importantly, notes will be taken and synthesised to create a resource of methodologies linked to relevant publications; as well as strengthening the links between researchers and gaining a better understanding of the current and future methodology needs of our community. Pre-conference and Post- conference continued support for this will be provided and promoted by the ERA group.
Wednesday, 31 July 2019
The concept of Smart Cities can be understood as a holistic approach to improve the level of development and management of the city in a broad range of services by using information and communication technologies.
It is common to recognize six axes of work in them: i) Smart Economy, ii) Smart People, iii) Smart Governance, iv) Smart Mobility, v) Smart Environment, and vi) Smart Living. In this tutorial we first focus on a capital issue: smart mobility. European citizens and economic actors need a transport system which provides them with seamless, high-quality door-to-door mobility. At the same time, the adverse effects of transport on the climate, the environment and human health need to be reduced. We will show many new systems based in the use of bio-inspired techniques to ease the road traffic flow in the city, as well as allowing a customized smooth experience for travelers (private and public transport).
This tutorial will then discuss on potential applications of intelligent systems for energy (like adaptive lighting in streets), environmental applications (like mobile sensors for air pollution), smart building (intelligent design), and several other applications linked to smart living, tourism, and smart municipal governance.
Cartesian Genetic Programming (CGP) is a well-known and respected form of Genetic Programming. Its generality means that it can be applied to a wide range of computational problems in many fields including Alife.
It uses a very simple integer address-based genetic representation of a program in the form of a directed graph. In various studies, CGP has been shown to be comparatively efficient to other GP techniques.
The classical form of CGP has undergone a number of developments which have made it more useful, efficient and flexible in various ways. These include self-modifying CGP (SMCGP), cyclic connections (recurrent-CGP), encoding artificial neural networks and automatically defined functions (modular CGP).
SMCGP uses functions that cause the evolved programs to change themselves as a function of time. Recurrent-CGP allows evolution to create programs which contain cyclic, as well as acyclic, connections. CGP encoded artificial neural networks represent a powerful training method for neural networks.
CGP has been applied successfully to a variety of real-world problems, such as digital circuit design, visual object recognition and classification. It has also been used to evolve programs that construct large structures or "organisms" using a process inspired by biological development.
CGP has a dedicated web site at www.cartesiangp.com
Modelling and simulation of complex problems has become an established ‘third pillar’ of science, complemen- tary to theory and experimentation. The multi-agent approach to modelling allows complex systems to be constructed in such as way as to add complexity from understanding at an individual level (i.e. a bottom-up approach). This approach is extremely powerful in a wide range of domains as diverse as computational biology to economics and physics. Whilst multi-agent modelling provides a natural and intuitive method to model systems the computational cost of performing large simulations is much greater than for top-down, system level alternatives.
In order for multi-agent modelling and simulation to be used as a tool for delivering excellent science, it is vital that simulation performance can scale, by targeting readily available computational resources effectively. Developed in UK since 2008, FLAME GPU provides this computational capacity by targeting readily available Graphics Processing Units capable of simulating many millions of interacting agents with performance which exceeds that of traditional CPU based simulators. FLAME GPU is an extended version of the FLAME (Flexible Large-scale Agent-based Modelling Environment) framework and is a mature and stable agent-based modelling simulation platform that enables modellers from various disciplines like economics, biology and social sciences to easily write agent-based models. Importantly it abstracts the complexities of the GPU architecture away from modellers to ensure that modellers can concentrate on writing models without the need to acquire specialist knowledge typically required to utilise GPU architectures.
This tutorial is aimed at the intermediate level. No knowledge of GPUs is required however basic knowledge multi agent modelling approaches is expected (i.e. formulating a problem as a set of individuals within a system) as well as understanding of XML document structure and basic programming ability.
By the end of the practical session, it is expected that the participants will understand how to write and execute a multi-agent model for FLAME GPU from scratch. Participants will leave with an appreciation of the key techniques, concepts, and algorithms which have been used.
Gene regulatory networks are a central mechanism in the regulation of gene expression in all living organisms’ cells. Their functioning is nowadays very well understood: they are based on the production of proteins enhanced or inhibited by other proteins from the inside and/or the outside of the cells. This produces complex dynamics with which cells from the same organism, with the exact same DNA, can form very different types (through cell differentiation) and have very different behaviors according to their types and positions.
This tutorial will first introduce the biology of these networks, from the genetic aspect to the dynamics of gene regulation. Then, biologically plausible models will be presented to highlight the complexity of dynamics gene regulatory networks can produce. Using that model, we will show how computational models can be designed so that a genetic algorithm can optimize the network efficiently. We will present a set of applications in which artificial gene regulatory networks are plugged into diverse virtual agents (artificial cells in an artificial embryogenesis context, direct connection to the sensors and effectors or high-level behavior regulation in others). Demos, showing the results obtained with this system, will be presented all along the tutorial.