[DMANET] PhD Position at Southampton (Reinforcement Learning)

PhD Studentship in Reinforcement Learning for Resource Allocation

URL: https://jobs.soton.ac.uk/Vacancy.aspx?ref=1116619FP

Project description

We welcome applications for a fully-funded PhD studentship in reinforcement learning for resource allocation within the Agents, Interaction and Complexity group at the University of Southampton (https://www.aic.ecs.soton.ac.uk).

The successful candidate will join the US/UK-funded Distributed Analytics and Information Science International Technology Alliance (DAIS ITA): https://dais-ita.org. This $100M research programme is developing the fundamental science underpinning future information systems, where intelligent software agents and humans will collaborate to efficiently collect, process and disseminate information across complex and dynamic networks.

The student will work closely within a team of four academic staff, four PhD students and a postdoctoral researcher on using techniques from artificial intelligence to develop agile resource allocation mechanisms. The overarching vision of this team is to create flexible and robust techniques for composing and executing workflows of computational tasks in highly-distributed and decentralised edge cloud systems.

The focus of this PhD studentship will be to develop novel reinforcement learning algorithms that take into consideration the unique characteristics of dynamic edge clouds. In particular, these algorithms need to be able to adapt to rapidly-changing circumstances, work well in previously unseen environments and deal with the presence of competing, often self-interested task requesters (which may be intelligent software agents or human users). The algorithms developed in this PhD will be directly applicable to a wide range of emerging application settings, including the Internet-of-Things (IoT), ad hoc networks for disaster response or intelligent transportation systems.

As part of this PhD, the student will advance the state of the art in artificial intelligence and will publish at internationally-leading conferences and journals. Furthermore, the project is a unique opportunity to collaborate with a wide range of academic, industrial and governmental institutions across the DAIS ITA programme. The student will be expected to attend and be an active contributor at regular project meetings that are held in both the UK and the US There will be opportunities to apply for internships with industry and government partners.

Please contact Dr Sebastian Stein (ss2@ecs.soton.ac.uk<mailto:ss2@ecs.soton.ac.uk>) or Prof Timothy Norman (t.j.norman@soton.ac.uk<mailto:t.j.norman@soton.ac.uk>) for further details.

Key facts

Entry requirements: first-class degree or equivalent in computer science or a related discipline.

Closing date: applications should be received no later than 12 April 2019. Earlier applications are encouraged. Later applications may be considered depending on the funds remaining.

Duration: three years (full-time)

Funding: full tuition fees (for UK/EU students) and a tax-free stipend of £16,000 per year. Applications from self-funded international students are welcome.

Assessment: 9-month and 18-month reports, viva voce and thesis examination

Start date: typically September, but an earlier start is possible.

Applying: www.southampton.ac.uk/postgraduate/pgstudy/howdoiapplypg.html<http://www.southampton.ac.uk/postgraduate/pgstudy/howdoiapplypg.html>


—
Dr Sebastian Stein
Associate Professor - Artificial Intelligence / Multi-Agent Systems

University of Southampton, UK
Electronics and Computer Science
Building 32, Room 4023

Tel: +44 (0) 23 8059 7645
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************