Fig 3 - uploaded by Michael Lones
Content may be subject to copyright.
Source publication
Biological organisms exist within environments in which complex, non-linear dynamics are ubiquitous. They are coupled to these environments via their own complex, dynamical networks of enzyme-mediated reactions, known as biochemical networks. These networks, in turn, control the growth and behaviour of an organism within its environment. In this pa...
Citations
... These computational dynamical systems (CDS) [37] include various kinds of recurrent neural network (RNN) and cellular automata, but also architectures motivated by the low-level biochemical networks that are directly exposed to evolution within biological systems. This includes our own work on artificial biochemical networks (ABNs) [38]. ...
... In previous work, we have looked at a number of variant ABN models, which include features such as dynamical nodal processes [51], self-modification [52], conservation laws [50], weak coupling between networks [53], and higher-order coupling [51]. We have found these architectures to be particularly useful for solving complex control problems, such as chaos control and legged robot locomotion [38], with different architectures being beneficial for different problems. For instance, self-modifying networks are useful when there is a requirement to switch dynamically between different behaviours. ...
... For instance, self-modifying networks are useful when there is a requirement to switch dynamically between different behaviours. Notably, we have found the use of discrete maps within network nodes to be beneficial across a diverse range of problems [38], [41]. ...
Parkinson's disease is a debilitating neurological condition that affects approximately 1 in 500 people and often leads to severe disability. To improve clinical care, better assessment tools are needed that increase the accuracy of differential diagnosis and disease monitoring. In this paper, we report how we have used evolutionary algorithms to induce classifiers capable of recognising the movement characteristics of Parkinson's disease patients. These diagnostically-relevant patterns of movement are known to occur over multiple time scales. To capture this, we used two different classifer architectures: sliding-window genetic programming classifiers, which model over-represented local patterns that occur within time series data, and artificial biochemical networks, computational dynamical systems that respond to dynamical patterns occurring over longer time scales. Classifiers were trained and validated using movement recordings of 49 patients and 41 age-matched controls collected during a recent clinical study. By combining classifiers with diverse behaviours, we were able to construct classifier ensembles with diagnostic accuracies in the region of 95%, comparable to the accuracies achieved by expert clinicians. Further analysis indicated a number of features of diagnostic relevance, including the differential effect of handedness and the over-representation of certain patterns of acceleration.
http://dx.doi.org/10.1109/TEVC.2013.2281532
The study of living systems—including those existing in nature, life as it could be, and even virtual life—needs consideration of not just traditional biology, but also computation and physics. These three areas need to be brought together to study living systems as cyber-bio-physical systems, as zoetic systems. Here I review some of the current work on assembling these areas, and how this could lead to a new Zoetic Science. I then discuss some of the significant scientific advances still needed to achieve this goal. I suggest how we might kick-start this new discipline of Zoetic Science through a program of Zoetic Engineering: designing and building living artefacts. The goal is for a new science, a new engineering discipline, and new technologies, of zoetic systems: self-producing far-from-equilibrium systems embodied in smart functional metamaterials with non-trivial meta-dynamics.
We introduce and experimentally demonstrate the utility of tag-based genetic regulation, a new genetic programming (GP) technique that allows programs to dynamically adjust which code modules to express.Tags are evolvable labels that provide a flexible mechanism for referencing code modules. Tag-based genetic regulation extends existing tag-based naming schemes to allow programs to “promote” and “repress” code modules in order to alter expression patterns. This extension allows evolution to structure a program as a gene regulatory network where modules are regulated based on instruction executions. We demonstrate the functionality of tag-based regulation on a range of program synthesis problems. We find that tag-based regulation improves problem-solving performance on context-dependent problems; that is, problems where programs must adjust how they respond to current inputs based on prior inputs. Indeed, the system could not evolve solutions to some context-dependent problems until regulation was added. Our implementation of tag-based genetic regulation is not universally beneficial, however. We identify scenarios where the correct response to a particular input never changes, rendering tag-based regulation an unneeded functionality that can sometimes impede adaptive evolution. Tag-based genetic regulation broadens our repertoire of techniques for evolving more dynamic genetic programs and can easily be incorporated into existing tag-enabled GP systems.
Evolutionary algorithms (EAs) are population-based metaheuristics, originally inspired by aspects of natural evolution. Modern varieties incorporate a broad mixture of search mechanisms, and tend to blend inspiration from nature with pragmatic engineering concerns; however, all EAs essentially operate by maintaining a population of potential solutions and in some way artificially 'evolving' that population over time. Particularly well-known categories of EAs include genetic algorithms (GAs), Genetic Programming (GP), and Evolution Strategies (ES). EAs have proven very successful in practical applications, particularly those requiring solutions to combinatorial problems. EAs are highly flexible and can be configured to address any optimization task, without the requirements for reformulation and/or simplification that would be needed for other techniques. However, this flexibility goes hand in hand with a cost: the tailoring of an EA's configuration and parameters, so as to provide robust performance for a given class of tasks, is often a complex and time-consuming process. This tailoring process is one of the many ongoing research areas associated with EAs. © Springer International Publishing AG, part of Springer Nature 2018. All rights reserved.
Evolutionary algorithms (EAs) are population-based metaheuristics, originally inspired by aspects of natural evolution. Modern varieties incorporate a broad mixture of search mechanisms, and tend to blend inspiration from nature with pragmatic engineering concerns; however, all EAs essentially operate by maintaining a population of potential solutions and in some way artificially 'evolving' that population over time. Particularly well-known categories of EAs include genetic algorithms (GAs), Genetic Programming (GP), and Evolution Strategies (ES). EAs have proven very successful in practical applications, particularly those requiring solutions to combinatorial problems. EAs are highly flexible and can be configured to address any optimization task, without the requirements for reformulation and/or simplification that would be needed for other techniques. However, this flexibility goes hand in hand with a cost: the tailoring of an EA's configuration and parameters, so as to provide robust performance for a given class of tasks, is often a complex and time-consuming process. This tailoring process is one of the many ongoing research areas associated with EAs.
Kaum eine Entwicklung innerhalb der Digitalisierung bietet so viel Stoff für Utopien und Dystopien wie die Künstliche Intelligenz (KI). Doch wer KI einsetzen will, muss die zugrunde liegenden Methoden kennen und bewerten können. Mit diesem Trendsonar wollen wir Ihnen nicht nur einen Überblick über aktuelle und künftige KI-Verfahren geben. Eine detaillierte Einschätzung der Zukunftsfähigkeit, des Reifegrades, der Marktdurchdringung, der Standardisierung und der Verfügbarkeit bieten Ihnen einen fundierten Einblick in den aktuellen Entwicklungsstand.
Gene regulatory networks (GRNs) are the fundamental mechanisms through which biological organisms control their growth, their dynamical behavior, their interaction with their environment, and which underlie much of the complexity in the biosphere. This chapter reviews current understanding of artificial gene regulatory network (AGRN), discussing what is known about their computational properties, detailing how they have been applied to computational problems, and speculating about how they may be used in the future. It discusses what is known about biological GRNs, and the implications this has for the design of AGRNs. The chapter presents the different motivations behind the development of AGRN models. It also discusses the modeling decisions that have to be made when developing AGRN models. AGRNs give the opportunity to explore analogous behaviors within a more general setting, which, in turn, might lead to a better understanding of the general properties of GRNs.
Levodopa is a drug that is commonly used to treat movement disorders associated with Parkinson's disease. Its dosage requires careful monitoring, since the required amount changes over time, and excess dosage can lead to muscle spasms known as levodopa-induced dyskinesia. In this work, we investigate the potential for using epiNet, a novel artificial gene regulatory network, as a classifier for monitoring accelerometry time series data collected from patients undergoing levodopa therapy. We also consider how dynamical analysis of epiNet classifiers and their transitions between different states can highlight clinically useful information which is not available through more conventional data mining techniques. The results show that epiNet is capable of discriminating between different movement patterns which are indicative of either insufficient or excessive levodopa.
Gene regulatory networks (GRNs) are the complex dynamical systems that orchestrate the activities of biological cells. In order to design effective therapeutic interventions for diseases such as cancer, there is a need to control GRNs in more sophisticated ways. Computational control methods offer the potential for discovering such interventions, but the difficulty of the control problem means that current methods can only be applied to GRNs that are either very small or that are topologically restricted. In this paper, we consider an alternative approach that uses evolutionary algorithms to design GRNs that can control other GRNs. This is motivated by previous work showing that computational models of GRNs can express complex control behaviours in a relatively compact fashion. As a first step towards this goal, we consider abstract Boolean network models of GRNs, demonstrating that Boolean networks can be evolved to control trajectories within other Boolean networks. The Boolean approach also has the advantage of a relatively easy mapping to synthetic biology implementations, offering a potential path to in vivo realisation of evolved controllers.
This paper describes the artificial epigenetic network (AEN), a recurrent connectionist architecture that is able to dynamically modify its topology in order to automatically decompose and solve dynamical problems. The approach is motivated by the behaviour of gene regulatory networks, particularly the epigenetic process of chromatin remodelling that leads to topological change and which underlies the differentiation of cells within complex biological organisms. We expected this approach to be useful in situations where there is a need to switch between different dynamical behaviours, and do so in a sensitive and robust manner in the absence of a priori information about problem structure. This hypothesis was tested using a series of dynamical control tasks, each requiring solutions which could express different dynamical behaviours at different stages within the task. In each case, the addition of topological self-modification was shown to improve the performance and robustness of controllers. We believe this is due to the ability of topological changes to stabilise attractors, promoting stability within a dynamical regime whilst allowing rapid switching between different regimes. Post hoc analysis of the controllers also demonstrated how the partitioning of the networks could provide new insights into problem structure.
