Fireflies flashing, birds flocking, fish schooling. Collective behaviours stop us in our tracks when we encounter them in nature, and we marvel at how such intricate displays come about. The groups appear to be solving tasks that the individuals constituting that group could not. How do dumb agents come together to act cleverly?
This need not be limited to the animal domain either. Just imagine that instead of seeing that flock of starlings on your morning stroll, you see a swarm of robots collectively solving their tasks. It may sound like sci-fi, but even in robotics, simple local interactions can give rise to new global behaviours. What can artificial agents teach us about the natural order, and how does inspiration from nature and brains improve the robotic performance?
And then there are neurons — yet another collection of otherwise "dumb agents" capable of more than the sum of its parts.
Natural agents. Artificial agents. Neural “agents”.
Against this wonderful scientific backdrop, we will also be joined by two creative groups to broaden our perspectives and interpret our theme through dance and through art.
Please join us for this year’s SWC and GCNU student symposium as we explore these ideas together. It will all take place online on October 8th from 10:30 BST.
The symposium will be hosted on Crowdcast. SIGN UP HERE!
For SWC students and staff
We are also planning some fabulous audience engagement. Think of your current work, let yourself be inspired how it might play out in collective actions, and stay tuned for details! There will be prizes.
Intro by the SWC/Gatsby Student Symposium Team
Iain Couzin, Max Planck Institute of Animal Behavior & University of Konstanz
Ivar Hagendoorn, a choreographer, photographer, and researcher
Sabine Hauert, University of Bristol
Viola Priesemann, Max Planck Institute for Dynamics and Self-organization
Lonneke Gordijn, Studio DRIFT, a pair of artists working on experimental sculptures, installations, and performances
Justin Werfel, Harvard University
Orit Peleg, University of Colorado Boulder & Santa Fe Institute
Adrienne Fairhall, University of Washington
In-person student presentations, pizzas, and drinks for the internal audience
*all timings in BST (UTC+1)
Talk titles and abstracts will be published below in due course. For now — acquaint yourself with our brilliant speakers by reading the following short profiles we have prepared.
Iain Couzin (website, @icouzin)
Iain Couzin is one of the directors of the Max Planck Institute of Animal Behavior in Konstanz and the chair of Biodiversity and Collective Behaviour at the University of Konstanz. His work is focussed on the fundamentals of collective behaviour: what influences swarm behaviour? How does the individual decide what to do within a swarm? To investigate this, he not only combines experiments and theory, but is also at the forefront of developing new technology: tracking several hundred fish simultaneously with a python-based system, or immersing fish in a virtual reality swarm are among the techniques developed in his lab.
Due to his interest in fundamentals, he has worked with many species from locusts to primates as well as the collective behaviour of single cells, and his team includes researchers from diverse backgrounds such as behavioural ecology, applied mathematics and physics.
The Geometry of Decision-Making
Choosing among spatially distributed options is a central challenge for animals, from deciding among alternative potential food sources or refuges, to choosing with whom to associate. Here, using an integrated theoretical and experimental approach (employing immersive Virtual Reality), with both invertebrate and vertebrate models—the fruit fly, desert locust and zebrafish—we consider the recursive interplay between movement and collective vectorial integration in the brain during decision-making regarding options (potential ‘targets’) in space. We reveal that the brain repeatedly breaks multi-choice decisions into a series of abrupt (critical) binary decisions in space-time where organisms switch, spontaneously, from averaging vectorial information among, to suddenly excluding one of, the remaining options. This bifurcation process repeats until only one option—the one ultimately selected—remains. Close to each bifurcation the ‘susceptibility’ of the system exhibits a sharp increase, inevitably causing small differences among the remaining options to become amplified; a property that both comes ‘for free’ and is highly desirable for decision-making. This mechanism facilitates highly effective decision-making, and is shown to be robust both to the number of options available, and to context, such as whether options are static (e.g. refuges) or mobile (e.g. other animals). In addition, we find evidence that the same geometric principles of decision-making occur across scales of biological organisation, from neural dynamics to animal collectives, suggesting they are fundamental features of spatiotemporal computation.
Ivar Hagendoorn (website, @IvarHagendoorn)
Ivar Hagendoorn is a choreographer, photographer and researcher. When not in the studio working on a new dance production, he can be found sitting behind his computer editing photos or writing a research paper.
His research applies insights from philosophy, cognitive neuroscience, experimental psychology, mathematics and sociology to the study of dance and choreography. In recent years, his research has shifted to the intersection of artificial intelligence, cognitive (neuro)science and philosophy. The focus is still on dance and creativity: "Ultimately I dream of developing a system capable of learning high level concepts, which can be used to generate dance phrases". On 8th October, he will go into detail about his project ‘Emergent Choreography’ that explores applications of complexity theory to dance. In particular, the tenets of complexity theory is that a central governing agent is not necessary for the emergence of intricate patterns or cooperative behaviour.
Self-Organized Choreography. From Interaction to Communication
A choreography can provisionally be defined as a set of instructions that determines the movements of one or multiple agents in space and time and their interactions. It follows that the more complex the choreography the longer the set of instructions. This in turn means that there is a limit to the amount of complexity that can be achieved in a given amount of time. To deal with increasing complexity choreographers focus either on the movements of the individual dancers or on collective movements. Complexity science offers a different paradigm to create a high degree of complexity by means of self-organization. First I will argue why complexity matters, why, in other words, complexity is an aesthetic value. Next I will outline some of the rules and principles that can be used to generate self-organizing patterns in the context of a dance performance. I will argue that when working with dancers one can use the full panoply of human cognitive capacities, from pattern recognition to the use of signs, to create intricate patterns that go beyond those seen in complex systems in nature and physics.
Sabine Hauert (website, @sabinehauert)
Sabine Hauert is Associate Professor (Reader) of Swarm Engineering at the University of Bristol in the UK. Her research focuses on making swarms for people, and across scales, from nanorobots for cancer treatment, to larger robots for environmental monitoring, or logistics. Before joining the University of Bristol, Sabine engineered swarms of nanoparticles for cancer treatment at MIT, and deployed swarms of flying robots at EPFL.
Sabine is also President and Co-founder of Robohub.org, and executive trustee of AIhub.org, two non-profits dedicated to connecting the robotics and AI communities to the public.
Swarms for people
As tiny robots become individually more sophisticated, and larger robots easier to mass produce, a breakdown of conventional disciplinary silos is enabling swarm engineering to be adopted across scales and applications, from nanomedicine to treat cancer, to cm-sized robots for large-scale environmental monitoring or intralogistics. This convergence of capabilities is facilitating the transfer of lessons learned from one scale to the other. Cm-sized robots that work in the 1000s may operate in a way similar to reaction-diffusion systems at the nanoscale, while sophisticated microrobots may have individual capabilities that allow them to achieve swarm behaviour reminiscent of larger robots with memory, computation, and communication. Although the physics of these systems are fundamentally different, much of their emergent swarm behaviours can be abstracted to their ability to move and react to their local environment. This presents an opportunity to build a unified framework for the engineering of swarms across scales that makes use of machine learning to automatically discover suitable agent designs and behaviours, digital twins to seamlessly move between the digital and physical world, and user studies to explore how to make swarms safe and trustworthy. Such a framework would push the envelope of swarm capabilities, towards making swarms for people.
Viola Priesemann (website, @violapriesemann)
Prof. Priesemann is a group leader at the Max Planck Institute for Dynamics and Self-Organization in Göttingen, Germany. She studies self-organizing systems, spreading processes, and how artificial and living networks can process information. Her work demonstrates how emergent dynamics of complex networks, such as reverberation, critical (or sub-critical) dynamics, and phase transitions can act as a substrate for information processing. During the COVID-19 pandemic, she has focused on creating much-needed and influential models of disease spread which quantified the effects of lockdowns and other interventions. Her remarkable scientific contributions and clarity in communicating with the public have been recognized by the Max Planck Society with the Communitas award.
Tuning dumb neurons to task processing - via homeostasis
Homeostatic plasticity plays a key role in stabilizing neural network activity. But what is its role in neural information processing? We showed analytically how homeostasis changes collective dynamics and consequently information flow - depending on the input to the network. We then studied how input and homeostasis on a recurrent network of LIF neurons impacts information flow and task performance. We showed how we can tune the working point of the network, and found that, contrary to previous assumptions, there is not one optimal working point for a family of tasks, but each task may require its own working point.
Lonneke Gordijn, Studio DRIFT (website, @studiodrift)
Lonneke Gordijn and Ralph Nauta founded studio DRIFT in 2006/7. Their work fuses hidden properties of nature with technology and mankind by combining for example nature and science fiction, knowledge and intuition. By doing so, they want to spark everyone’s curiosity and open people’s eyes to everything we take for granted, to wonder how everyday objects work and to reconnect us with our planet. To make us realize that with our current technology we can try to approach, but never fully cover how nature works because of its complexity. Their work is very unique because of their interdisciplinary collaborations with scientists, universities, engineers and many more people. According to Ralph, persistence is key to making everything happen and Lonneke the world is one big exhibition if you only care to look.
About how we can see a way forward by observing nature.
Justin Werfel (website)
Justin Werfel leads the Designing Emergence Laboratory at Harvard University. His research interests are in the understanding and design of collective behavior in complex and emergent systems, with work in areas including swarm robotics, social insect behavior, evolutionary theory, engineered molecular nanosystems, and educational technology. His work has been featured by numerous national and international media, highlighted among Science's "top 10 scientific achievements of 2014", and denounced by a former assistant secretary of the US Treasury as "an enemy of the human race."
Collective Construction in Natural and Artificial Swarms
Natural systems provide both puzzles to unravel and demonstrations of what's possible. The natural world is full of complex systems of dynamically interchangeable, individually unreliable components that produce effective and reliable outcomes at the group level. A complementary goal to understanding the operation of such systems is that of being able to engineer artifacts that work in a similar way. One notable type of collective behavior is collective construction, epitomized by mound-building termites, which build towering, intricate mounds through the joint activity of millions of independent and limited insects. The artificial counterpart would be swarms of robots designed to build human-relevant structures. I will discuss work on both aspects of the problem, including studies of cues that individual termite workers use to help direct their actions and coordinate colony activity, and development of robot systems that build user-specified structures despite limited information and unpredictable variability in the process. These examples illustrate principles used by the insects and show how they can be applied in systems we create.
Orit Peleg (website, @oritpeleg)
Setting up GoPros in a forest or working with 10k subjects at once is nothing unusual for Orit Peleg. As an Assistant Professor at the University of Colorado Boulder and External Faculty at the Santa Fe Institute, she seeks to understand how individual organisms dynamically harness the emergent capabilities of the collective to guard themselves against environmental fluctuations. Combining field work with controlled experiments and mathematical modelling, her lab studies a range of fascinating behaviours in disordered living systems, such as collective ventilation in honeybee hives and synchronous flashing of fireflies. Their work demonstrates how short-range interactions between locally perceptive individuals can result in spontaneous reorganisation of their macro-environment in a way that enhances their collective survival and, at least in the case of fireflies, mesmerises their beholder.
Physical Computation in Insect Swarms
Our world is full of living creatures that must share information to survive and reproduce. As humans, we easily forget how hard it is to communicate within natural environments. So how do organisms solve this challenge, using only natural resources? Ideas from computer science, physics and mathematics, such as energetic cost, compression, and detectability, define universal criteria that almost all communication systems must meet. We use insect swarms as a model system for identifying how organisms harness the dynamics of communication signals, perform spatiotemporal integration of these signals, and propagate those signals to neighboring organisms. In this talk I will focus on two types of communication in insect swarms: visual communication, in which fireflies communicate over long distances using light signals, and chemical communication, in which bees serve as signal amplifiers to propagate pheromone-based information about the queen’s location.
Adrienne Fairhall (website, @alfairhall)
A theoretical physicist by training, Adrienne Fairhall has carved out her niche in computational neuroscience by applying dynamical systems ideas to study the true nature of the neural code. Through collaborations with experimentalists like Rafael Yuste and Vanessa Ruta, her lab investigates the principles of neural computation in organisms as diverse as hydra, songbirds, insects, and mammals. Coincidentally, Adrienne's proudest piece of work, revealed in a recent Brain Inspired episode, is also one that transcends multiple systems, identifying fractional order differentiation as a good model for spike-rate adaptation in fruit fly and rat neurons alike (Fairhall et al. 2001b, Lundstrom et al. 2008). However, even though adaptive coding has been a major theme in her career, Adrienne's work is by no means limited to individual "dumb agents" but also considers their role at the level of networks and, indeed, animal behaviour. And if you find yourself wondering how Adrienne can be such a kick-ass scientist, mentor, and mum all at once, treat yourself to a series of stories by inspiring women scientists that Adrienne has compiled as proof that you can, in fact, have it all.
Reverse engineering Hydra
Hydra is an extraordinary creature. Continuously replacing itself, it can live indefinitely, performing a stable repertoire of reasonably sophisticated behaviors. This remarkable stability under plasticity may be due to the uniform nature of its nervous system, which consists of two apparently noncommunicating nerve net layers. We use modeling to understand how active muscles and biomechanics interact with neural activity to shape Hydra behavior. We will discuss our findings and thoughts on how this simple nervous system may self-organize to produce purposeful behavior.
SWC/GCNU Student Symposium Team
The Student Symposium is organised jointly by PhD students of the Sainsbury Wellcome Centre for Neural Circuits and Behaviour and the Gatsby Computational Neuroscience Unit of University College London. This is the fourth instalment of an annual discussion-based event that aims to bring together neuroscience researchers from the UK and abroad to engage with current and future problems in neuroscience.