Richard E. Turner
Professor of Machine Learning
Works on probabilistic machine learning and its use for weather and earth-system prediction, including the Aardvark Weather and Aurora foundation models.
Richard E. Turner's lab at Cambridge works on probabilistic & causal inference, generative & foundation models and large language models. 41 papers since 2023, 29 at top venues, mostly NeurIPS and ICLR. 10 PhD students. 19 alumni, most often to Microsoft and Citadel. Applications close .
At a glance
- papers, 2023–2026
- 11/ year
- at T1 venues
- 71%
- PhD students
- 10
- MS students
- 0
- alumni listed
- 19
Papers by yearfrom DBLP, 2023–2026
- 2 papers2026 · T1 2 · collecting
- 13 papers2025 · T1 9
- 16 papers2024 · T1 12
- 10 papers2023 · T1 6
T1 venuesOtherVenues: NeurIPS, ICLR, ICML, AISTATS, ICCV
Research focuspaper titles that use each topic's words
- Probabilistic & causal inference5 papers
- Generative & foundation models5 papers
- Large language models5 papers
- weather1 paper
- probabilistic machine0 papers
Paper list
From DBLP.
20262 papers
202513 papers
- Nat.A foundation model for the Earth system (opens in a new tab)
- AISTATS (T1)Bayesian Circular Regression with von Mises Quasi-Processes (opens in a new tab)
- NeurIPS (T1)Distributional Training Data Attribution: What do Influence Functions Sample? (opens in a new tab)
- Trans. Mach. Learn. Res.Efficient Few-Shot Continual Learning in Vision-Language Models (opens in a new tab)
- Nat.End-to-end data-driven weather prediction (opens in a new tab)
- NeurIPS (T1)Estimating Interventional Distributions with Uncertain Causal Graphs through Meta-Learning (opens in a new tab)
- ICML (T1)Gridded Transformer Neural Processes for Spatio-Temporal Data (opens in a new tab)
- ICLR (T1)Influence Functions for Scalable Data Attribution in Diffusion Models (opens in a new tab)
- ICLR (T1)Linear Transformer Topological Masking with Graph Random Features (opens in a new tab)
- ICML (T1)Position: Probabilistic Modelling is Sufficient for Causal Inference (opens in a new tab)
- NeurIPS (T1)Purifying Shampoo: Investigating Shampoo's Heuristics by Decomposing its Preconditioner (opens in a new tab)
- Trans. Mach. Learn. Res.Tighter sparse variational Gaussian processes (opens in a new tab)
202416 papers
- NeurIPS (T1)A Generative Model of Symmetry Transformations (opens in a new tab)
- NeurIPS (T1)Approximately Equivariant Neural Processes (opens in a new tab)
- MLHCBeyond Clinical Trials: Using Real World Evidence to Investigate Heterogeneous, Time-Varying Treatment Effects (opens in a new tab)
- ICML (T1)Can We Remove the Square-Root in Adaptive Gradient Methods? A Second-Order Perspective (opens in a new tab)
- NeurIPS (T1)Fearless Stochasticity in Expectation Propagation (opens in a new tab)
- AISTATS (T1)Identifiable Feature Learning for Spatial Data with Nonlinear ICA (opens in a new tab)
- AABIImplicitly Bayesian Prediction Rules in Deep Learning (opens in a new tab)
- AABIIn-Context In-Context Learning with Transformer Neural Processes (opens in a new tab)
- NeurIPS (T1)LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language (opens in a new tab)
- NeurIPS (T1)Noise-Aware Differentially Private Regression via Meta-Learning (opens in a new tab)
- NeurIPS (T1)On conditional diffusion models for PDE simulations (opens in a new tab)
- AISTATS (T1)Optimising Distributions with Natural Gradient Surrogates (opens in a new tab)
202310 papers
- ICLR (T1)Autoregressive Conditional Neural Processes (opens in a new tab)
- CoLLAsComparing the Efficacy of Fine-Tuning and Meta-Learning for Few-Shot Policy Imitation (opens in a new tab)
- Trans. Mach. Learn. Res.Differentially private partitioned variational inference (opens in a new tab)
- ICCV (T1)First Session Adaptation: A Strong Replay-Free Baseline for Class-Incremental Learning (opens in a new tab)
- ICLR (T1)FiT: Parameter Efficient Few-shot Transfer Learning for Personalized and Federated Image Classification (opens in a new tab)
- NeurIPS (T1)Geometric Neural Diffusion Processes (opens in a new tab)
- Trans. Mach. Learn. Res.Improving Continual Learning by Accurate Gradient Reconstructions of the Past (opens in a new tab)
- NeurIPS (T1)Kronecker-Factored Approximate Curvature for Modern Neural Network Architectures (opens in a new tab)
- Trans. Mach. Learn. Res.On the Efficacy of Differentially Private Few-shot Image Classification (opens in a new tab)
- NeurIPS (T1)PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solvers (opens in a new tab)
Where alumni went19 listed, 67% to industry
Industry 8Academia 4
Most often to
- Microsoft
- Citadel
Also to
- Harvard University
- the University of Sydney 2013 2017
- University of Cambridge
Apply to CambridgeMichaelmas 2027
PhD in Computer Science
- Round
- Funding round (8 Dec 2026)
- Opens
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- USMIT
Joshua Tenenbaum
- Computational neuroscience & cognition
- CHEPFL
Volkan Cevher
- Learning theory & optimization
- GBImperial
Yingzhen Li
- Probabilistic & causal inference
- USUIUC
Nan Jiang
- Learning theory & optimization
- CNPeking
Liwei Wang
- Learning theory & optimization
- USUT Austin
Hyeji Kim
- Learning theory & optimization
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- DBLP
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Cambridgecloses