Rethinking Bayesian active learning

Lead Research Organisation: University of Oxford


Brief description of the context of the research including potential impact
From drug design to policymaking, machine learning is transforming the way we make decisions by allowing us to draw inferences from complex data. At a fundamental level this relies on the quality of the data we use. Yet, compared with intense attention on methods for learning from a given dataset, little research has dealt with what data to acquire in the first place. Our work focuses on general-purpose methods for intelligent data acquisition, building on the principled foundations of information theory. Advances in these methods stand to significantly boost the practical utility of machine learning in society.

Aims and objectives
We aim to develop methods for Bayesian active learning, in which an algorithm seeks the most informative data for training a predictive model. Our plan comprises three steps.

First, we will rethink the notion of information gain with which the algorithm decides what data to acquire. Whereas the field has historically focused on gathering information about abstract model parameters, we will design methods that focus on the predictions we ultimately want to make with the model. This change in the acquisition objective can drastically improve the relevance of the data the algorithm acquires.

Second, we will argue for the importance of capitalising on all data at hand in order to inform data-acquisition decisions. When we want our model to make predictions on a particular task, incorporating additional task-generic data into the model does not just improve the model's baseline predictive performance: it can enable better-targeted data acquisition too.

Third, we will introduce techniques for non-myopic data acquisition. The predominant approach to active learning is to make a sequence of greedy, myopic data-acquisition decisions, where each decision is assumed to be the last. We will design data-acquisition algorithms that can reason about the total information that can be gathered using a given acquisition budget. This could yield a further improvement in the usefulness of the datasets we acquire in active learning.

Novelty of the research methodology
Our work combines empirical investigation with theoretical exposition.

Alignment to EPSRC's strategies and research areas
This research relates to a number of EPSRC research areas: artificial intelligence technologies; digital signal processing; image and vision computing; information systems; statistics and applied probability.

Planned Impact

AIMS's impact will be felt across domains of acute need within the UK. We expect AIMS to benefit: UK economic performance, through start-up creation; existing UK firms, both through research and addressing skills needs; UK health, by contributing to cancer research, and quality of life, through the delivery of autonomous vehicles; UK public understanding of and policy related to the transformational societal change engendered by autonomous systems.

Autonomous systems are acknowledged by essentially all stakeholders as important to the future UK economy. PwC claim that there is a £232 billion opportunity offered by AI to the UK economy by 2030 (10% of GDP). AIMS has an excellent track record of leadership in spinout creation, and will continue to foster the commercial projects of its students, through the provision of training in IP, licensing and entrepreneurship. With the help of Oxford Science Innovation (investment fund) and Oxford University Innovation (technology transfer office), student projects will be evaluated for commercial potential.

AIMS will also concretely contribute to UK economic competitiveness by meeting the UK's needs for experts in autonomous systems. To meet this need, AIMS will train cohorts with advanced skills that span the breadth of AI, machine learning, robotics, verification and sensor systems. The relevance of the training to the needs of industry will be ensured by the industrial partnerships at the heart of AIMS. These partnerships will also ensure that AIMS will produce research that directly targets UK industrial needs. Our partners span a wide range of UK sectors, including energy, transport, infrastructure, factory automation, finance, health, space and other extreme environments.

The autonomous systems that AIMS will enable also offer the prospect of epochal change in the UK's quality of life and health. As put by former Digital Secretary Matt Hancock, "whether it's improving travel, making banking easier or helping people live longer, AI is already revolutionising our economy and our society." AIMS will help to realise this potential through its delivery of trained experts and targeted research. In particular, two of the four Grand Challenge missions in the UK Industrial Strategy highlight the positive societal impact underpinned by autonomous systems. The "Artificial Intelligence and data" challenge has as its mission to "Use data, Artificial Intelligence and innovation to transform the prevention, early diagnosis and treatment of chronic diseases by 2030". To this mission, AIMS will contribute the outputs of its research pillar on cancer research. The "Future of mobility" challenge highlights the importance the autonomous vehicles will have in making transport "safer, cleaner and better connected." To this challenge, AIMS offers the world-leading research of its robotic systems research pillar.

AIMS will further promote the positive realisation of autonomous technologies through direct influence on policy. The world-leading academics amongst AIMS's supervisory pool are well-connected to policy formation e.g. Prof Osborne serving as a Commissioner on the Independent Commission on the Future of Work. Further, Dr Dan Mawson, Head of the Economy Unit; Economy and Strategic Analysis Team at BEIS will serve as an advisor to AIMS, ensuring bidirectional influence between policy objectives and AIMS research and training.

Broad understanding of autonomous systems is crucial in making a society robust to the transformations they will engender. AIMS will foster such understanding through its provision of opportunities for AIMS students to directly engage with the public. Given the broad societal importance of getting autonomous systems right, AIMS will deliver core training on the ethical, governance, economic and societal implications of autonomous systems.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/S024050/1 01/10/2019 31/03/2028
2416587 Studentship EP/S024050/1 01/10/2020 30/09/2024 Frederick Bickford-Smith