Peter J. Bentley

Peter J. Bentley
University College London | UCL · Department of Computer Science

Ph.D. B.Sc. (Hons)

About

330
Publications
73,140
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
7,686
Citations
Citations since 2016
48 Research Items
1917 Citations
2016201720182019202020212022050100150200250300
2016201720182019202020212022050100150200250300
2016201720182019202020212022050100150200250300
2016201720182019202020212022050100150200250300
Additional affiliations
January 1997 - present
University College London

Publications

Publications (330)
Book
There's a hidden science that affects every part of your life. You are fluent in its terminology of email, WiFi, social networking, and encryption. You use its results when you make a telephone call, access the Internet, use any factory-produced product, or travel in any modern car. The discipline is so new that some prefer to call it a branch of...
Book
Full-text available
Evolutionary Design by Computers is a collection of essays that describe recent research into "evolutionary" computing where computers mimic the strategies of biological evolution to solve problems in architecture, engineering, art and artificial life. Peter Bentley's excellent introduction to the current state of evolutionary design quickly direct...
Conference Paper
Full-text available
Collaboration is an essential aspect of human interaction. Despite being mutually beneficial to everyone involved, it often fails due to behaviour differences as individuals process information, form opinions, and interact with each other, especially when their task contains uncertainty. Thus, to understand collaboration on noisy problems effective...
Conference Paper
Full-text available
Nature has spent billions of years perfecting our genetic representations , making them evolvable and expressive. Generative machine learning offers a shortcut: learn an evolvable latent space with implicit biases towards better solutions. We present SOLVE: Search space Optimization with Latent Variable Evolution, which creates a dataset of solutio...
Preprint
Full-text available
We present a temporally extended variation of the successor representation, which we term t-SR. t-SR captures the expected state transition dynamics of temporally extended actions by constructing successor representations over primitive action repeats. This form of temporal abstraction does not learn a top-down hierarchy of pertinent task structure...
Article
Full-text available
We have a wide breadth of computational tools available today that enable a more ethical approach to the study of human cognition and behavior. We argue that the use of computer models to study evolving ecosystems provides a rich source of inspiration, as they enable the study of complex systems that change over time. Often employing a combination...
Preprint
Full-text available
Teams are central to human accomplishment. Over the past half-century, psychologists have identified the Big-Five cross-culturally valid personality variables: Neuroticism, Extraversion, Openness, Conscientiousness, and Agreeableness. The first four have shown consistent relationships with team performance. Agreeableness (being harmonious, altruist...
Chapter
Real-world design problems are a messy combination of constraints, objectives, and features. Exploring these problem spaces can be defined as a Multi-Criteria Exploration (MCX) problem, whose goals are to produce a set of diverse solutions with high performance across many objectives, while avoiding low performance across any objectives. Quality-Di...
Conference Paper
Full-text available
Constrained optimization problems can be difficult because their search spaces have properties not conducive to search, e.g., multimodality, discontinuities, or deception. To address such difficulties, considerable research has been performed on creating novel evolutionary algorithms or specialized genetic operators. However, if the representation...
Preprint
Real-world design problems are a messy combination of constraints, objectives, and features. Exploring these problem spaces can be defined as a Multi-Criteria Exploration (MCX) problem, whose goals are to produce a set of diverse solutions with high performance across many objectives, while avoiding low performance across any objectives. Quality-Di...
Preprint
Full-text available
Constrained optimization problems can be difficult because their search spaces have properties not conducive to search, e.g., multimodality, discontinuities, or deception. To address such difficulties, considerable research has been performed on creating novel evolutionary algorithms or specialized genetic operators. However, if the representation...
Article
Full-text available
Real-world design problems are a messy combination of constraints, objectives, and features. Exploring these problem spaces can be defined as a Multi-Criteria Exploration (MCX) problem, whose goals are to produce a set of diverse solutions with high performance across many objectives, while avoiding low performance across any objectives. Quality-Di...
Conference Paper
Full-text available
Machine Learning has the potential to discover new correlations between energy usage in apartments and variables such as seasonality, apartment location, size, efficiency and details of those staying in the apartments, thus helping apartments to become more sustainable and helping those who stay in them to use less energy. The biggest impedance to...
Article
Full-text available
We describe the Intelligent Autopilot System (IAS), a fully autonomous autopilot capable of piloting large jets such as airliners by learning from experienced human pilots using Artificial Neural Networks. The IAS is capable of autonomously executing the required piloting tasks and handling the different flight phases to fly an aircraft from one ai...
Conference Paper
Full-text available
According to Gray's Reinforcement Sensitivity Theory (RST), individuals have differing sensitivities to rewards and punishments, which in turn affect their behaviours. The behavioural inhibition system (BIS) is associated with sensitivity to punishment while the behavioural activation system (BAS) is associated with sensitivity to reward. In this w...
Preprint
Full-text available
We propose PiNet, a generalised differentiable attention-based pooling mechanism for utilising graph convolution operations for graph level classification. We demonstrate high sample efficiency and superior performance over other graph neural networks in distinguishing isomorphic graph classes, as well as competitive results with state of the art m...
Conference Paper
We can talk about learning optimisation in terms of three biological processes: evolution, development and learning. It has been argued that all three are necessary for intelligence to emerge. Together, they shape the brain through regressive and progressive plasticity. In this paper, we explored the effects of structural plasticity on learning in...
Article
Evolution provides a creative fount of complex and subtle adaptations that often surprise the scientists who discover them. However, the creativity of evolution is not limited to the natural world: Artificial organisms evolving in computational environments have also elicited surprise and wonder from the researchers studying them. The process of ev...
Preprint
Full-text available
Maintaining the ability to fire sparsely is crucial for information encoding in neural networks. Additionally, spiking homeostasis is vital for spiking neural networks with changing numbers of weights and neurons. We discuss a range of network stabilisation approaches, inspired by homeostatic synaptic plasticity mechanisms reported in the brain. Th...
Conference Paper
Full-text available
Maintaining the ability to fire sparsely is crucial for information encoding in neural networks. Additionally, spiking homeostasis is vital for spiking neural networks with changing numbers of weights and neurons. We discuss a range of network stabilisation approaches, inspired by homeostatic synaptic plasticity mechanisms reported in the brain. Th...
Conference Paper
Full-text available
Change is inevitable in this fast-moving world. As the environment and people’s needs continuously change, so must the project. In our previous work, we developed an agent-based model of human collaboration that incorporates individual personalities. In this work, we applied a genetic algorithm to select the optimal personality combinations of a te...
Conference Paper
Colorectal cancer (CRC) is the second most common tumour in the world (Bray, 2018). It has been proposed that morbidity and mortality could be mitigated by screening methods that identify key genetic mutations in the DNA of a patient’s biosample (Traverso, 2002). However, for this to work, a theoretical understanding of the most likely mutations th...
Conference Paper
Full-text available
Collaboration is fundamental to our society, but how should we best build teams? We investigate by applying optimisation to an agent-based model of collaboration. The model takes inspiration from particle swarm optimisation, abstracting a shared goal as a shared optimisation task, and modelling the personality differences in team members as strateg...
Preprint
We propose an end-to-end deep learning learning model for graph classification and representation learning that is invariant to permutation of the nodes of the input graphs. We address the challenge of learning a fixed size graph representation for graphs of varying dimensions through a differentiable node attention pooling mechanism. In addition t...
Preprint
Graph classification is a significant problem in many scientific domains. It addresses tasks such as the classification of proteins and chemical compounds into categories according to their functions, or chemical and structural properties. In a supervised setting, this problem can be framed as learning the structure, features and relationships betw...
Conference Paper
Spiking neural networks, thanks to their sensitivity to the timing of the inputs, are a promising tool for unsupervised processing of spatio-temporal data. However, they do not perform as well as the traditional machine learning approaches and their real-world applications are still limited. Various supervised and reinforcement learning methods for...
Conference Paper
This work proposes a method for predicting the internal mechanisms of individual agents using observed collective behaviours by multi-agent reinforcement learning (MARL). Since the emergence of group behaviour among many agents can undergo phase transitions, and the action space will not in general be smooth, natural evolution strategies were adopt...
Conference Paper
Full-text available
In this work we present IHDNs: an original model of computation for the simulation of interacting, dynamic, multi-scale systems. We show that a novel message passing mechanism that operates across layers of abstraction in hierarchical dynamic networks is effective in expressing the complex dependencies of living systems. Using a conventional comput...
Conference Paper
Full-text available
Intestinal glands in the small intestine and colon, or intestine crypts, are an important example of tissue homeostasis regulated by the extracellular environment. The crypts are invaginated structures made of a layer of cells that help absorb nutrients from passing food. However, they are continuously worn away by this process and are being contin...
Article
Full-text available
Biological evolution provides a creative fount of complex and subtle adaptations, often surprising the scientists who discover them. However, because evolution is an algorithmic process that transcends the substrate in which it occurs, evolution's creativity is not limited to nature. Indeed, many researchers in the field of digital evolution have o...
Conference Paper
Full-text available
Spiking neural networks have been previously used to perform tasks such as object recognition without supervision. One of the concerns relating to the spiking neural networks is their speed of operation and the number of iterations necessary to train and use the network. Here, we propose a biologically plausible model of a spiking neural network wh...
Conference Paper
Full-text available
We introduce the Intelligent Autopilot System (IAS) which is capable of autonomous navigation and landing of large jets such as airliners by observing and imitating human pilots using Artificial Neural Networks and Learning by Imitation. The IAS is a potential solution to the current problem of Automatic Flight Control Systems of being unable to pe...
Conference Paper
Full-text available
We introduce the PseudoGravity tool, an automated social media system that establishes a social media presence in the area of interest of a target audience, identifies target users that are open to connect, engages with them, and elicits a complex response and time investment from them. In this work, we use Twitter as the social media platform and...
Conference Paper
Full-text available
We introduce the Intelligent Autopilot System (IAS) which is capable of autonomous landing, and go-around of large jets such as airliners under severe weather conditions. The IAS is a potential solution to the current problem of Automatic Flight Control Systems of being unable to autonomously handle flight uncertainties such as severe weather condi...
Conference Paper
We propose an extension to the capabiliti es of the Intelligent Autopilot System (IAS) from our previou s work, to be able to learn handling emergencies by observing and imitating human pilots. The IAS is a potential solution to th e current problem of Automatic Flight Control Systems of bein g unable to handle flight uncertainties, and the need to...
Conference Paper
Biological systems have become highly significant for traditional computer architectures as examples of highly complex self-organizing systems that perform tasks in parallel with no centralized control. However, few researchers have compared the suitability of different computing approaches for the unique features of Artificial Immune Systems (AIS)...
Conference Paper
An Intelligent Autopilot System (IAS) that can learn piloting skills by observing and imitating expert human pilots is proposed. IAS is a potential solution to the current problem of Automatic Flight Control Systems of being unable to handle flight uncertainties, and the need to construct control models manually. A robust Learning by Imitation appr...
Article
Full-text available
Design Computing and Cognition (DCC'14) - Volume 30 Issue 2 - Sean Hanna, Peter Bentley
Article
This paper evaluates different clustering methods used within the Artificial Ecosystem Algorithm (AEA). The AEA is designed to take advantage of highly distributed computer architectures and adapt to changing problems. In the AEA a problem is first decomposed into its relative sub-components: they then evolve to form solution building blocks that f...
Article
Full-text available
Predictions which invoke evolutionary mechanisms are hard to test. Agent-based modeling in artificial life offers a way to simulate behaviors and interactions in specific physical or social environments over many generations. The outcomes have implications for understanding adaptive value of behaviors in context. Pain-related behavior in animals is...
Article
Predictions which invoke evolutionary mechanisms are hard to test. Agent-based modeling in artificial life offers a way to simulate behaviors and interactions in specific physical or social environments over many generations. The outcomes have implications for understanding adaptive value of behaviors in context. Pain-related behavior in animals is...
Article
Full-text available
Natural ecosystems exhibit complex dynamics of interacting species. Man-made ecosystems exhibit similar dynamics, and in the case of mobile app stores, can be said to perform optimization as developers seek to maximize app downloads. This work aims to understand stability and instability within app store dynamics and how it affects fitness. The inv...
Conference Paper
We extend and adapt the Artificial Ecosystem Algorithm (AEA), by applying it to the dynamic redistribution of bicycles in London’s Santander Cycle scheme. Just as an ecosystem comprises many separate components that adapt to form a single synergistic whole, the AEA uses a bottom up approach to build a solution. A problem is decomposed into relative...
Chapter
Full-text available
iStethoscope Pro is the first piece of software (an "App") produced for iOS devices, which enabled users to exploit their smartphones, music players, or tablets as stethoscopes. The software exploits the built-in microphone (and supports externally added microphones) and performs real-time amplification and filtering to enable heart sounds to be he...
Conference Paper
An ecosystem inspired algorithm that aims to take advantage of highly distributed computer architectures is proposed. The motivation behind this work is to grasp the phenomenal properties of ecosystems and use them for large-scale real-world problems. Just as an ecosystem comprises many separate components that adapt together to form a single syner...
Article
Full-text available
Systemic computation (SC) is a bio-inspired computational paradigm designed to model the behaviour of natural systems and processes. It adopts a holistic view, meaning that apart from a sum of its constituents, the definition of a system should also include the interaction of its elements. SC implies an unconventional massively parallel computer ar...
Article
Full-text available
Mobile applications (apps) are software developed for use on mobile devices and made available through app stores. App stores are highly competitive markets where developers need to cater to a large number of users spanning multiple countries. This work hypothesizes that there exist country differences in mobile app user behavior and conducts one o...
Conference Paper
Full-text available
An ecosystem inspired algorithm that aims to take advantage of highly distributed computer architectures is proposed. Our motivation is to grasp the phenomenal properties of ecosystems and use them for large-scale real-world problems. Just as an ecosystem comprises of many separate components that adapt together to form a single synergistic whole,...
Conference Paper
Full-text available
Real world machine learning, where data is sampled continuously, may in theory be classifiable into distinct and unchanging categories but in practice the classification becomes non-trivial because the nature of the background noise continuously changes. Applying distinct and unchanging categories for data ignores the fact that for some application...
Conference Paper
Agent-Based Models are used to model dynamic systems such as stock markets, societies, and complex biological systems that are difficult to model analytically using partial differential equations. Many agent-based modeling software are designed for serial von-Neumann computer architectures. That limits the speed and scalability of these systems. Sy...
Article
Full-text available
Many cancers are aneuploid. However, the precise role that chromosomal instability plays in the development of cancer and in the response of tumours to treatment is still hotly debated. Here, to explore this question from a theoretical standpoint we have developed an agent-based model of tissue homeostasis in which to test the likely effects of who...
Article
Full-text available
Being able to engineer a set of components and their corresponding environmental conditions such that target entities emerge as the result of self-assembly remains an elusive goal. In particular, understanding how to exploit physical properties to create self-assembling systems in three dimensions (in terms of component movement) with symmetric and...
Conference Paper
Full-text available
App stores are one of the most popular ways of providing content to mobile device users today. But with thousands of competing apps and thousands new each day, the problem of presenting the developers' apps to users becomes non-trivial. There may be an app for everything, but if the user cannot find the app they desire, then the app store has faile...
Conference Paper
Systemic Computation is an unconventional paradigm which defines a model of natural behavior and implies a massively parallel computer architecture. It is designed to be a computational paradigm for natural systems and processes modeling. Existing software implementations have been too limited in terms of performance, flexibility and programmabilit...
Article
Abstract One of the practical challenges facing the creation of self-assembling systems is being able to exploit a limited set of fixed components and their bonding mechanisms. The method of staging divides the self-assembly process into time intervals, during which components can be added to, or removed from, an environment at each interval. Stagi...
Article
In this paper we describe a methodology for heart sound classification and results obtained at PASCAL Classifying Heart Sounds Challenge. The results of competing methodologies are shown. The approach has two steps: segmentation and classification of heart sounds. We also describe the data collection procedure.
Chapter
Throughout nature, in both the inorganic and organic realms, complex entities emerge as a result of self-assembly from decentralised components governed by simple rules. Natural self-assembly is dictated by the morphology of the components and the environmental conditions they are subjected to, as well as the physical and chemical properties of the...
Article
Full-text available
You’re waiting at the station for your train and you glance at the electronic poster next to you. It notices that you’re looking at it, and from your gaze it works out what you would most like to see. The display changes to show you new brands of mobile phone, and then changes again to show handheld computers as it notices your gaze flicker. You gl...
Data
Full-text available
In mobile app ecosystems, an app can behave like a virus. Once downloaded, it may cause its user to recommend that app to friends who then may download the app and "infect" other friends. Epidemics occur when a small number of downloads causes a snowballing effect that results in a massive number of downloads (and consequently, a rich developer). T...
Conference Paper
Full-text available
In mobile app ecosystems, an app can behave like a virus. Once downloaded, it may cause its user to recommend that app to friends who then may download the app and "infect" other friends. Epidemics occur when a small number of downloads causes a snowballing effect that results in a massive number of downloads (and consequently, a rich developer). T...