Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Collective Learning in Multi-Agent Systems


   Department of Computer Science

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Dr O Akanyeti  No more applications being accepted  Funded PhD Project (European/UK Students Only)

About the Project

Lead supervisor: Otar Akanyeti (Intelligent Robotics Group, IMPACS, Aberystwyth University; [Email Address Removed]); Co-supervisor: Sarah Dalesman (Aquatic, Behavioural & Evolutionary Biology, IBERS, Aberystwyth University; [Email Address Removed])
Project
Studies on bacteria swarms, ant colonies, fish shoals, and bird flocks clearly show that a group can achieve much more than the sum of its parts. Simple rules of interactions among neighbors, which require no centralized control or detailed knowledge of the environment can lead to enhanced sensing, movement and decision-making capabilities. However, individuals are also capable of achieving more than simply following predefined sets of rules. Division of labor in insect colonies or honey bee foraging with a myriad of dance signals and behavioral cues are just few examples that cannot be fully explained in terms of simple individuals alone.

It turns out that the ability to learn and the heterogeneity of the group (individual variations among group members in terms of skills, personality and experience) can lead to enhanced group performances in dynamic scenarios. Until now, we have had a rudimentary idea on how learning occurs at the group level, how information flows among group members, and what learning strategy each member should follow to maximize the information gain of the group. These are fundamental questions on collective learning (e.g. https://doi.org/10.1371/journal.pone.0015505), which are highly relevant for a wide range of fields including neuroscience, swarm robotics and search and optimization algorithms.

The PhD project will develop new cooperative control strategies for underwater robots using the principles of self-organization and collective learning. Information gained from biological experiments on collective learning in fish will be used to refine computational models. The main goal of this project is to discover new bio-inspired learning strategies for multi-agent systems by employing a combination of biological and robotics experiments as well as computer simulations.

This PhD studentship provides an opportunity to gain world-class expertise in computational science and animal behaviour. Akanyeti (lead supervisor) has an active research program in bio-inspired underwater robotics. Dalesman (co-supervisor) has extensive experience in measuring learning and memory, animal personality and physiology. IMPACS houses one of the UK’s leading research groups on robotics that design systems for space, aerial, terrestrial and water applications. IBERS provides state of the art flexible aquarium facilities, providing ideal space for the biological elements.

Through the cross-disciplinary training provided you will develop a wide range of skills applicable to robotics, computational modelling and behavioural research. You will be expected to present your findings at major international conferences and receive training in making your research accessible through publishing scientific papers and public engagement.
Eligibility
This is an interdisciplinary project and the student will be part of the Institute of Mathematics, Physics and Computer Science and Institute of Biological, Environmental and Rural Sciences. We are looking for someone motivated and have a strong passion for science. A good degree (2:1 minimum) and interest in biology and computer science is desired. No prior knowledge is required.
Equality and Diversity
Aberystwyth University is committed to promoting equality and diversity, and endeavours to be inclusive, valuing the diversity of its staff, students and community. Expressions of interest from women would be particularly welcome, as well as from other suitably-qualified individuals from a wide variety of backgrounds. Where appropriate, all reasonable adjustments are made to enable appointees to effectively carry out their duties.
Funding
The PhD studentship will cover tuition fees at the home rate (for UK and EU students only) and provide a tax-free stipend at the standard RCUK rate. The length of the studentship is 3 years (full-time).
Application Process
Applications should be submitted online to the Postgraduate Admissions Office using the following link: https://www.aber.ac.uk/en/postgrad/apply/. Required documents are:
1) A completed Postgraduate Application Form, plus two references submitted by the deadline.
2) A PhD proposal of up to 1,000 words where you expand on your experience and interests and describe why you are a good candidate for this research studentship. please enter the lead supervisor name under “Project title applied for”.
Location
Aberystwyth is a small, friendly and attractive town on the west coast of Wales, UK, with low crime rate and many opportunities for outdoor activities (e.g. hiking, canoeing, horse riding and surfing). The town is set in rural surroundings and within easy reach of Pembrokeshire Coast and Snowdonia National Parks. The student will be based on the modern Penglais campus with a pleasant site overlooking the town and Cardigan Bay.
Contact Details
Informal enquiries should be made to Dr. Otar Akanyeti at [Email Address Removed] or 01970622537.

Welsh details of the advert are available on request.

 About the Project