Brooks Paige

Email me: b.paige@ucl.ac.uk

I am an associate professor in machine learning at the University College London AI Centre.

I’m also a Turing fellow at the Alan Turing Institute, and a statistical ambassador for the Royal Statistical Society.

I’m interested in developing interpretable machine learning models which complement human expertise, rather than attempt to replace it. For a high-level overview on some recent-ish work, here’s a video of a talk I gave at the PROBPROG conference in Boston in September, 2018.

Previously I was a Turing research fellow based at the Alan Turing Institute; before that I was a DPhil student at University of Oxford, working with Frank Wood. My thesis focused on sequential Monte Carlo methods, and their application to general-purpose inference in probabilistic programs. A cliff-notes version is available as this short talk.

Brooks: a photo

Current highlights

Seasonal Arctic sea ice forecasting with probabilistic deep learning, recently published in Nature Communications, is the result of an exciting collaboration with the British Antarctic Survey and the Alan Turing Institute, among others. We find that a probabilistic deep learning model, trained on a mix of historical and simulated data, can provide state-of-the-art prediction of Arctic sea ice extent. You can learn more about this from this short video with Tom Andersson, who led the study.

The Molecule Chef, and its extention to arbitrary synthetic routes represented as directed acyclic graphs, introduce a new deep generative models for molecules, which produces novel molecules by selecting combinations of easily available precursors which will react together when run through a reaction predictor. This is follow-up work to our earlier papers on molecular search using discrete deep generative models, and modeling chemical reactions while respecting chemical constraints.

An introduction to probabilistic programming, co-written with Jan-Willem, Hongseok, and Frank, is a textbook-length tutorial on developing probabilistic programming systems and that can perform automatic Bayesian inference on a large class of models.

Publications

This list tends to be slightly outdated; please also refer to my google scholar profile.

Working papers

Willets, M., & Paige, B. (2021). I Don’t Need u: Identifiable Non-Linear ICA Without Side Information. [preprint]

van de Meent, J.-W., Paige, B., Yang, H., & Wood, F. (2019). An Introduction to Probabilistic Programming. In revision, Foundations and Trends in Machine Learning (FTML). [preprint]

Journals and peer-reviewed conferences

Andersson, T.R., Hosking, J.S., Pérez-Ortiz, M., Paige, B., Elliott, A., Russell, C., Law, S., Jones, D.C., Wilkinson, J., Phillips, T. and Byrne, J. (2021). Seasonal Arctic sea ice forecasting with probabilistic deep learning. Nature Communications, 12(1), 1–12. [link]

Shi, Y., Paige, B., Siddharth, N., & Torr, P.H.S. (2021). Relating by contrasting: A data-efficient framework for multimodal DGMs. In International Conference on Learning Representations (ICLR). [link]

Camuto, A., Willetts, M., Paige, B., Holmes, C., & Roberts, S. (2021). Learning Bijective Feature Maps for Linear ICA. In Proceedings of The 24th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 130:3655-3663. [link].

Paige, B., Bell, J., Bellet, A., Gascón, A., & Ezer, D. (2021). Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores. In Journal of Computational Biology, 28 (5), 435–451 [preprint]

Bradshaw, J., Paige, B., Kusner, M., Segler, M.H.S., & Hernández-Lobato, J.M. (2020). Barking up the right tree: an approach to search over molecule synthesis DAGs In Advances in Neural Information Processing Systems (NeurIPS), 33:6852–6866.

Mollaysa, A. & Paige, B., & Kalousis, A. (2020). Goal-directed Generation of Discrete Structures with Conditional Generative Models. In Advances in Neural Information Processing Systems (NeurIPS), 33:21923–21933.

Clymo, J., Manukian, H., Fijalkow, N., Gascón, A., & Paige, B. (2020). Data Generation for Neural Programming by Example. In Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics, PMLR (AISTATS), 108:3450-3459. [link]

Bradshaw, J., Paige, B., Kusner, M., Segler, M.H.S., & Hernández-Lobato, J.M. (2019). A Model to Search for Synthesizable Molecules. In Advances in Neural Information Processing Systems (NeurIPS), 32:7935–7947. [link]

Shi, Y., Siddharth, N., Paige, B., & Torr, P.H.S. (2019). Variational Mixture-of-Experts Autoencoders for Multi-Modal Deep Generative Models. In Advances in Neural Information Processing Systems (NeurIPS), 32:15692–15703. [link]

Law, S., Paige, B., & Russell, C. (2019). Take a Look Around: Using Street View and Satellite Images to Estimate House Prices. In ACM Transactions on Intelligent Systems and Technology (TIST), 10, 5, Article 54. [link] [arXiv]

Bradshaw, J., Kusner, M., Paige, B., Segler, M.H.S., & Hernández-Lobato, J.M. (2019). A Generative Model for Electron Paths. In International Conference on Learning Representations (ICLR). [preprint]

Esmaeili, B., Wu, H., Jain, S., Bozkurt, A., Siddharth, N., Paige, B., Brooks, D.H., Dy, J., van de Meent, J.-W. (2019). Structured Disentangled Representations. In Proceedings of Machine Learning Research (AISTATS), PMLR 89:2525-2534. [link]

Janz, D., van der Westhuizen, J., Paige, B., Kusner, M., & Hernández-Lobato, J.M. (2018). Learning a Generative Model for Validity in Complex Discrete Structures. In International Conference on Learning Representations (ICLR). [link]

Siddharth, N.*, Paige, B.*, van de Meent, J.-W.*, Desmaison, A., Wood, F., Goodman, N.D., Kohli, P., & Torr, P.H.S. (2017). Learning Disentangled Representations with Semi-Supervised Deep Generative Models. In Advances in Neural Information Processing Systems (NIPS), 30:5925–5935. [link]

Kusner, M.J.*, Paige, B.*, & Hernández-Lobato, J.M. (2017). Grammar Variational Autoencoder. In Proceedings of the 34th International Conference on Machine Learning, PMLR 70: 1945–1954. [link] [video]

Schuster, I., Strathmann, H., Paige, B., & Sejdinovic, D. (2017). Kernel Sequential Monte Carlo. In European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), LNCS 10534: 390–409. [preprint]

Paige, B. (2016). Automatic Inference for Higher-Order Probabilistic Programs. DPhil thesis. [link]

Paige, B., Sejdinovic, D., & Wood, F. (2016). Super-sampling with a Reservoir. In Proceedings of the 32nd Annual Conference on Uncertainty in Artificial Intelligence, UAI 32: 567–576. [link]

Paige, B., & Wood, F. (2016). Inference Networks for Sequential Monte Carlo in Graphical Models. In Proceedings of the 33rd International Conference on Machine Learning, JMLR W&CP 48: 3040-3049. [link]

Rainforth, T., Naesseth, C., Lindsten, F., Paige, B., van de Meent, J.-W., Doucet, A., & Wood, F. (2016). Interacting Particle Markov Chain Monte Carlo. In Proceedings of the 33rd International Conference on Machine Learning, JMLR W&CP 48: 2616-2625. [link]

van de Meent, J.-W., Paige, B., Tolpin, D., & Wood, F. (2016). Black-box policy search with probabilistic programs. In Proceedings of the 19th International Conference on Artificial Intelligence and Statistics (AISTATS), JMLR: W&CP 51: 1195-1204. [link]

Tolpin, D., van de Meent, J.-W., Paige, B., & Wood, F. (2015). Output-Sensitive Adaptive Metropolis-Hastings for Probabilistic Programs. In European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), LNCS 9285: 311-326. [preprint]

Paige, B., Wood, F., Doucet, A., & Teh, Y.W. (2014). Asynchronous Anytime Sequential Monte Carlo. In Advances in Neural Information Processing Systems (NIPS) 27: 3410-3418. [link] [video]

Paige, B., & Wood, F. (2014). A Compilation Target for Probabilistic Programming Languages. In Proceedings of the 31st International Conference on Machine Learning, JMLR W&CP 32(1): 1935-1943. [link] [video]

Shababo, B.*, Paige, B.*, Pakman, A. & Paninski, L. (2013). Bayesian inference and online experimental design for mapping neural microcircuits. In Advances in Neural Information Processing Systems (NIPS) 26: 1304-1312. [link]

Miklós, I., Paige, B., & Ligeti, P. (2006). Efficient Sampling of transpositions and inverted transpositions for Bayesian MCMC. In Proceedings of WABI 2006, LNBI 4175: 174-185.

Workshop publications, refereed abstracts, technical reports

Lamb, G., & Paige, B. (2020). Bayesian Graph Neural Networks for Molecular Property Prediction. At Machine Learning for Molecules Workshop @ NeurIPS 2020. [link]

Pappu, A. & Paige, B. (2020). Making Graph Neural Networks Worth It for Low-Data Molecular Machine Learning. At Machine Learning for Molecules Workshop @ NeurIPS 2020. [link]

Mollaysa, A. & Paige, B., & Kalousis, A. (2020). Conditional generation of molecules from disentangled representations. At Machine Learning for Molecules Workshop @ NeurIPS 2020. [link]

Marasoiu, M., Islam, S., Church, L., Lucero, M., Paige, B., Petricek, T. (2018). Stories of storytelling about UK’s EU funding. In Proceedings of the 2nd European Data and Computational Journalism Conference. [proceedings]

León-Villagrá, P., Islam, S., Lucero, M., Paige, B., Petricek, T. (2018). You guessed it! Reflecting on preconceptions and exploring data without statistics. In Proceedings of the 2nd European Data and Computational Journalism Conference. [proceedings]

Siddharth, N., Paige, B., Desmaison, A., van de Meent, J.-W., Wood, F., Goodman, N.D., Kohli, P., Torr, P.H.S. (2016). Inducing Interpretable Representations with Variational Autoencoders. Presented at Interpretable Machine Learning in Complex Systems, NIPS 2016 Workshop. [arXiv]

Janz, D., Paige, B., Rainforth, T., van de Meent, J.-W., Wood, F. (2016). Probabilistic structure discovery in time series data. Presented at Towards an Artificial Intelligence for Data Science, NIPS 2016 Workshop. [arXiv]

Paige, B., Wood, F. (2015). Inference networks for graphical models. Presented at Advances in Approximate Bayesian Inference, NIPS 2015 Workshop. [link]

Tolpin, D., Paige, B., van de Meent, J.-W., Wood, F. (2015). Path finding under uncertainty through probabilistic inference. Presented at 5th Workshop on Planning and Learning, ICAPS 2015. [arXiv]

van de Meent, J.-W., Paige, B., & Wood, F. (2014). Tempering by Subsampling. ArXiv e-prints, 1401.7145. [preprint]

Paige, B., Zhang, X., Forde, J., & Wood, F. (2012). Perspective Inference for Eye-to-Eye Videoconferencing: Empirical Evaluation Tools and Data. Presented at 7th Annual Machine Learning Workshop, New York Academy of Sciences.

* denotes joint first authorship.