Coordination strategies in multiagent reinforcement learning


An exciting opportunity to apply for a fully funded PhD position  in the College of Arts, Technology and Environment, UWE Bristol. The studentship will be funded by the the Computer Science Research Centre, UWE Bristol:

Ref: 2324-JAN-CATE03.

The expected start date of these studentships is 1 April 2024.

The closing date for applications is 20 November 2023.

Studentship details

This is an exciting opportunity to study for a fully funded PhD student to join the Multiagent Systems and Applications team that is part of the Computer Science Research Centre.

The aim of the project is to explore novel coordination approaches for multiagent reinforcement learning. Reinforcement learning is a learning approach for sequential decision-making processes where an agent learns to maximise a cumulative reward by trial and error. Although substantial progress has been made in the last years with novel reinforcement learning algorithms and tools, several challenges remain open, particularly in multiagent reinforcement learning environments. In these environments, agents have individual views of the environment and seek to maximise individual or collective rewards, depending on the cooperative or competitive nature of their goals. As part of the learning process, agents might coordinate their actions with others to improve their returns.

The project will focus on the evaluation and exploration of decentralised co- ordination models capturing the influence of agents’ actions on one other using local inputs to achieve a general/global perspective. The appointed researcher is expected to develop proof-of-concepts of environments, algorithms, and coordination models in this respect, as required. To build these models, it is foreseen that graph theory frameworks will be explored alongside game theory for analysis and design of coordination mechanisms. Such mechanisms should consider the natural local inefficiencies, the heterogeneity of agents and their relationships together with the system scale and dynamics.

Key expectations include:

  • Study the state-of-the-art models for decentralised local coordination with advantages and limitations.
  • Design novel models addressing some of the limitations observed.
  • Develop algorithms and environments required to evaluate the models under consideration for study.
  • Identify key scenarios and evaluate performance of analysed models.
  • Write manuscripts presenting key contributions.

For an informal discussion about the studentship, please email the supervisory team: Dr Marco Perez Hernandez ( or Dr Mehmet Aydin (


The studentship is available from 1 April 2024 for a period of three years, subject to satisfactory progress and includes a tax exempt stipend, which will be £18,622 (2023/24) per annum.

In addition, full-time tuition fees will be covered for up to three years.


The opportunity is open for UK/EU and International students.

  • Applicants must have a good honours BSc degree (2.1 or equivalent) in an appropriate discipline such as Computer Science, Engineering, Mathematics or similar.
  • Applicants whose first language is not English require a recognised English language qualification.
  • Applicants must be curious individuals, highly motivated to pursue a research career.
  • Good programming skills are essential.
  • Reasonable familiarity with formal analysis and mathematics is desirable.
  • Ability to work independently.

How to apply

Please submit your application online. When prompted, use the reference number 2324-JAN-CATE03.

Supporting documentation: You will need to upload your research proposal, all your degree certificates and transcripts and a recognised English language qualification (if required).

References: You will need to provide details of two referees as part of your application.

Closing date

The closing date for applications is Monday 20 November 2023.

Further information

The date of the interview will be communicated to the shortlisted candidates at least one week in advance. If you have not heard from us by 16 December 2023, we thank you for your application but, on this occasion, you have not been successful.

You may also be interested in