Skip to main content
Menu

Fully Funded 4-year Doctoral Studentship/Amazon Fellowship Joint with Amazon and the EPSRC CDT in Autonomous Intelligent Machines & Systems (AIMS)

AIMS logo, part of Oxford's Engineering Department Research Group

Fully Funded 4-year Doctoral Studentship/Amazon Fellowship

Joint with Amazon and the EPSRC CDT in Autonomous Intelligent Machines & Systems (AIMS)

Note: This studentship is fully funded.

Supervisor(s): Ingmar Posner and Jakob Foerster

Start Date: October 2025

Autonomous systems powered by artificial intelligence will have a transformative impact on economy, industry and society as a whole. Our mission is to train cohorts with both theoretical, practical and systems skills in autonomous systems - comprising machine learning, robotics, sensor systems and verification- and a deep understanding of the cross-disciplinary requirements of these domains. Industrial partnerships have been and will continue to be at the heart of AIMS, shaping its training and ensuring the delivery of Oxford’s world-leading research in autonomous systems to a wide variety of sectors, including smart health, transport, finance, energy and extreme environments. Given the broad importance of autonomous systems, AIMS provides training on the ethical, governance, economic and societal implications of autonomous systems. For more information regarding the AIMS programme, see our web pages at: aims.robots.ox.ac.uk.

Title: Versatile, Task-Driven World Models for Safe and Efficient Robot Learning

Abstract

The ability of robots to act and interact in unstructured, real-world environments remains a central challenge in robotics research. Decision-making, particularly in long-horizon tasks, is a key area of interest. While reinforcement learning (RL) has shown promise, its limitations in terms of data efficiency, safety, and interpretability have restricted its application to simulated environments or relatively simple real-world tasks.

However, recent advances in model-based learning offer a potential solution. In particular, by learning a world model from interaction data and optimising a policy within this model, agents can explore the state space (or observation space if coupled with a suitable generative model) in imagination without requiring additional real-world samples. However, a critical question is how to collect data that is relevant for learning policies for specific downstream tasks. Information Maximization approaches, while effective in reducing uncertainty about the environment, may not prioritise data that is relevant for the task or tasks at hand. Similarly, effectively curating the world model such that it becomes suitable for an expanding set of down-stream tasks remains an open challenge. This project aims to investigate data-driven, task-oriented methods for guiding data gathering for both model- and policy improvement in model-based learning. We will explore these topics using traditional benchmark datasets in machine learning as well as real-world robotics applications.

Award Value 

The studentship covers the full course fees plus a stipend (tax-free maintenance grant).

Eligibility

Prospective candidates will be judged according to how well they meet the following criteria:

        Applicants are normally expected to be predicted or have achieved a first-class or strong upper second-class undergraduate degree with honours (or equivalent international qualifications), as a minimum, in computer science, engineering, physics, mathematics, statistics or other related disciplines. A previous master's qualification is not required.

·        Excellent English written and spoken communication skills

Candidates will also need to demonstrate a broad interest in some or all of the four AIMS themes:

  •        machine learning, as a unifying core
  •         robotics & vision
  •     cyber-physical systems (e.g. sensor networks)
  •      control & verification

The deadline for applying is Tuesday 28th January 2025.  Candidates are therefore recommended to apply as soon as possible to and to inform wendy.adams@eng.ox.ac.uk when they have done so.

If you have any technical questions about the DPhil Studentship, please email wendy.adams@eng.ox.ac.uk

Please quote AIMS-AMAZON-2025 in the studentship reference box.

aims.robots.ox.ac.uk

Further details and how to apply can be found here: Autonomous Intelligent Machines and Systems (EPSRC CDT) | University of Oxford