HOR-INTERN-2020-05: Traffic scheduling in home IoT networks using deep reinforcement learning.

HOR-INTERN-2020-05: Traffic scheduling in home IoT networks using deep reinforcement learning

To apply, please complete the Internships Application Form.

Closing date

Applications must be made before 30th April 2020 and are assessed on an ongoing basis.

Location

University of Cambridge, Computer Laboratory reporting to Dr. Diana Andreea Popescu; this programme is administered by Horizon at the University of Nottingham as part of the EPSRC DADA project (EP/R03351X/1) involving Cambridge, Edinburgh and Imperial as well as Nottingham.

Proposed dates

Internships run for up to 12 weeks and are available for successful candidates to start anytime. Please indicate your availability on the application form, where requested.

Overview

Network traffic scheduling is traditionally done through customised policy rules installed in routers. In this project, we would like to explore a different approach for traffic scheduling that takes into account the dynamic variations of traffic using deep reinforcement learning. Starting from the open-source simulator (https://bitbucket.org/sandeep_chinchali/aaai18_deeprlcell/src) which works for IoT cellular networks, we would like to adapt the simulator for home IoT network scenarios.

Who should apply?

Ideal applicants should be studying for an undergraduate or postgraduate degree in mathematics, statistics, computer science or related discipline and have experience with a Python and machine learning.

Applicants must outline their area of interest, experience and skills. Information regarding Horizon, DADA and University of Cambridge, Computer Laboratory may be found at http://www.horizon.ac.uk/, https://www.horizon.ac.uk/project/defence-against-dark-artefacts/ and https://www.cst.cam.ac.uk/.

Required skills

  • Python

  • Knowledge of machine learning

  • Basic knowledge of networking

  • Ability to work independently as well as part of a team.

Desirable skills

  • Knowledge of reinforcement learning

  • Experience with academic writing.

Eligibility and financial aspects

These are full-time internships for up to 12 weeks, for postgraduate and undergraduate students, a casual wage of £425 per week will be available, and this may be subject to tax deductions depending on the successful candidate’s circumstances and stipend stipulations.

In general, PhD students from The University of Nottingham are able to apply on the understanding they suspend their stipend, this is due to the nature of the funding source. For International students a Visa must be in place, covering the duration of the internship.

Informal enquiries

Informal enquiries may be made to Dr. Diana Andreea Popescu diana.popescu@cl.cam.ac.uk

However applications should be made using the following web link. Applications to this email address will not be accepted.

To apply, please complete the Internships Application Form.