Powered by Possible

Our researchers pursue novel research in the lab and use the best ideas to power solutions that deliver real impact.

Fundamental Research

Pursuing novel avenues in key AI research areas including computer vision, natural language processing and machine learning
Applied Research

Exploring how AI can be implemented in real-world situations, testing prototypes and developing applications in key areas such as robotics and human-computer interaction
AI Impact

Studying the impact of AI on the world, including ethics and explainability, as well as the potential for using AI to tackle challenges in areas such as climate change and human rights

We bring the best of academia and industry
together to push the state of the art forward.

Bold & Impactful

More than 40 PhDs working on the biggest challenges in AI
Open & Collaborative

Accelerating discoveries means working across teams and in collaboration with an international network of academic fellows and AI experts
Deep expertise

Our researchers come from a wide range of disciplines and have the opportunity to work closely with the wider academic community, providing access to the latest ideas and talent
Benevolent

Researching the impact of AI and working towards models that are unbiased, fair, accurate and secure

Featured research

In collaboration with Yoshua Bengio, we developed N-Beats, a time-series forecasting model with a deep neural architecture that outperforms the leaders on a range of different competition datasets, including M4, M3 and TOURISM.

In addition to offering improved accuracy, N-Beats does not need to be fine-tuned, is fast to train and deploy, and easy to interpret.

Read the paper

Into the Unknown

Our fundamental research teams work on problems with unknown solutions, where research is ongoing or where we want to push the state-of-the-art.

Computer Vision

Vision for embodied agents, recognizing new tasks and leveraging knowledge from previous tasks, learning from minimal supervision, learning shared representations
Natural Language Processing

Document understanding, natural language interface, text representation and generalization, systematic generalisation, langage grounding
Machine Learning

Machine learning theory, reusability, robustness and explainability, scalability