The LCFI website uses cookies only for anonymised website statistics and for ensuring our security, never for tracking or identifying you individually. To find out more, and to find out how we protect your personal information, please read our privacy policy.

Read our privacy policy

Dismiss

Adrian Weller

Programme Director

Biography

Adrian Weller is Programme Director for Trust and Society at the CFI. He is also a Principal Research Fellow in Machine Learning at Cambridge, and Programme Director for AI at The Alan Turing Institute, the UK national institute for data science and AI, where he is a Turing Fellow leading work on safe, ethical and trustworthy AI.

His interests span AI, its commercial applications, and helping to ensure beneficial outcomes for society. He serves on several boards including the Centre for Data Ethics and Innovation. He is co-director of the European Laboratory for Learning and Intelligent Systems (ELLIS) programme on Human-centric Machine Learning, a member of the UNESCO Ad Hoc Expert Group on the Ethics of AI and a member of the IEEE Standards Working Group on Explainable AI.

Previously, Adrian held senior roles in finance.

Back to people

Adrian Weller

Biography

Challenges for Transparency

Comment: Humans can’t escape accountability for decisions made by artificial intelligence

Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty

Should the government play a role in reducing algorithmic bias?

Evaluating and Aggregating Feature-based Model Explanations

The Technology

Stochastic Flows and Geometric Optimization on the Orthogonal Group

How to talk to robots - The road to a people powered, AI-enabled future with Tabitha Goldstaub

AI and data governance from principles to practice: auto insurance. Workshop report September 2018

Always on my mind

Beyond Reasonable Doubt: Improving Fairness in Budget-Constrained Decision Making Using Confidence Thresholds

Rethinking Attention with Performers

TibGM: A Transferable and Information-Based Graphical Model Approach for Reinforcement Learning

Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch

“Explaining” machine learning reveals policy challenges

Machine Learning Explainability for External Stakeholders

Discovering interpretable representations for both deep generative and discriminative models

Machine Learning Explainability for External Stakeholders

Getting a CLUE: A method for explaining uncertainty estimates

An Algorithmic Framework for Positive Action

On the Fairness of Causal Algorithmic Recourse

An Algorithmic Framework for Positive Action

Machine Learning and the Meaning of Equal Treatment

Exploring Properties of the Deep Image Prior

CWY Parametrization: a Solution for Parallelized Optimization of Orthogonal and Stiefel Matrices

One-network Adversarial Fairness

Human Perceptions of Fairness in Algorithmic Decision Making: A Case Study of Criminal Risk Prediction

Beyond Distributive Fairness in Algorithmic Decision Making: Feature Selection for Procedurally Fair Learning

The Sensitivity of Counterfactual Fairness to Unmeasured Confounding

Explainable Machine Learning in Deployment

You shouldn't trust me: Learning models which conceal unfairness from multiple explanation methods

Getting a CLUE: A Method for Explaining Uncertainty Estimates

Counterfactual Accuracies of Alternative Models

Leveraging Data Science To Combat COVID-19: A Comprehensive Review

Motivations and Risks of Machine Ethics

Improving Interpretability in Medical Imaging Diagnosis using Adversarial Training

Robust Inverse Reinforcement Learning under Transition Dynamics Mismatch

Parametrization: a Solution for Parallelized Optimization of Orthogonal and Stiefel Matrices

Orthogonal Over-Parameterized Training

Improving Interpretability in Medical Imaging Diagnosis using Adversarial Training

Now You See Me (CME): Concept-based Model Extraction

On the Fairness of Causal Algorithmic Recourse

Stochastic Flows and Geometric Optimization on the Orthogonal Group

Now You See Me (CME): Concept-based Model Extraction

Sub-Linear Memory: How to Make Performers SLiM

From Parity to Preference-based Notions of Fairness in Classification

Debiasing a First-order Heuristic for Approximate Bi-level Optimization

An Empirical Study on Learning Fairness Metrics for COMPAS Data with Human Supervision

Sub-Linear Memory: How to Make Performers SLiM

Debiasing a First-order Heuristic for Approximate Bi-level Optimization

Iterative Teaching by Label Synthesis. In Neural Information Processing Systems

DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learning

Effects of Uncertainty on the Quality of Feature Importance Estimates

Purple Feed: Identifying High Consensus News Posts on Social Media

Learning with Hyperspherical Uniformity

Blind Justice: Fairness with Encrypted Sensitive Attributes

Iterative Teaching by Label Synthesis

Learning with Hyperspherical Uniformity

Adversarial Graph Embeddings for Fair Influence Maximization over Social Networks

Effects of Uncertainty on the Quality of Feature Importance Estimates

Orthogonal Over-Parameterized Training

Rethinking Attention with Performers

Mapping Intelligence: Requirements and Possibilities

Uncertainty as a Form of Transparency: Measuring, Communicating, and Using Uncertainty

Trust and Transparency

Faith and AI