PhD Studentship in Computational Neuroscience
Biocomputation Research Group
Centre for AI and Robotics Research
School of Physics, Engineering and Computer Science
University of Hertfordshire, UK

Informal enquiries: Prof Volker Steuber, UH ([email protected]); Dr Arnd 
Roth, UCL ([email protected])

Application deadline: 14 November 2025

Studentship: approximately £20,700 per annum bursary plus payment of the 
student fees. Applicants from inside and outside the UK are eligible.

Proposed Project: Reinforcement Learning and Microzones in the Cerebellum

The crystalline structure of the cerebellar cortex has inspired the development 
of  theories and computational models of learning in the cerebellum. In these 
computational models, learning is typically based on long-term depression (LTD) 
between parallel fibres and Purkinje cells in cerebellar cortex. More recently, 
however, plasticity has been shown to be present at different types of 
cerebellar synapses, and recent observations of reward signals conveyed by both 
mossy fibres and climbing fibres to the cerebellar cortex indicate that the 
roles of the instructive signals for cerebellar plasticity are more complex 
than assumed by classic theories. The presence of reward signals suggests that 
the cerebellum may be involved in reinforcement learning, by predicting the 
consequences of different actions. A further level of complexity arises from 
the existence of alternating cerebellar microzones that have been shown to 
contribute in different ways to cerebellar learning.

The proposed PhD project will extend a previously developed detailed network 
model of the cerebellum. Simulations of the network model will be used to 
investigate the implications of reward signals and microzones for cerebellar 
learning. The work will contribute to a better understanding of learning in the 
cerebellum with potential applications in machine learning and neurorobotics.

Applicants should have a keen interest in computational neuroscience and an 
excellent first degree or MSc in computer science, neuroscience, AI, machine 
learning, physics, maths, biology, medicine or a related discipline.

The project involves a collaboration between the Biocomputation Research Group 
in the Centre for AI and Robotics Research at the University of Hertfordshire 
(Volker Steuber) and the Wolfson Institute for Biomedical Research at UCL (Arnd 
Roth) and will require regular visits to London. The Centre for AI and Robotics 
Research and the University of Hertfordshire provide a very stimulating 
environment, offering a large Doctoral School with many specialised and 
interdisciplinary seminars as well as general training and researcher 
development opportunities. The University is situated in Hatfield, in the green 
belt just north of London.

Research in Computer Science at the University of Hertfordshire has been 
recognised as excellent in REF 2021, with 90% of the research submitted and all 
of the research impact rated as internationally excellent or world leading.

Information on the application procedure and an application form can be found 
at: 
https://www.herts.ac.uk/study/schools-of-study/physics-engineering-and-computer-science/engineering-and-computer-science/research-in-engineering-and-computer-science/the-phd-programme-in-computer-science


Volker Steuber
Professor of Computational Neuroscience
Biocomputation Research Group
School of Physics, Engineering and Computer Science
University of Hertfordshire
College Lane, Hatfield, AL10 9AB, UK
biocomputation.herts.ac.uk<http://biocomputation.herts.ac.uk>

-- 
Systems Neuroscience mailing list: www.systemsneuroscience.eu
To post, send email to [email protected]
--- 
You received this message because you are subscribed to the Google Groups 
"Systems Neuroscience" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/systems-neuroscience/CWXP265MB396008FF3F4FB74404D921CDA4F6A%40CWXP265MB3960.GBRP265.PROD.OUTLOOK.COM.

Reply via email to