CSML@Cam - Talks schedule

We meet every week on Friday at 14:00. If you are interested in giving a talk, email Connor (cpd32@cam.ac.uk) to organise a date and time.


2022-01-21 - TBA - Toni Karvonen

Past talks

Michaelmas Term 2021

2021-12-10 - Bayesian Learning via Neural Schrodinger Follmer Flow - Francisco Vargas Palomo
2021-11-26 - Kronecker-based network generation and analysis - Thomas Gaskin
2021-11-19 - Satellite remote sensing (Part II) - Andrea Marinoni
2021-11-12 - Learning based multiscale modeling - Burigede Liu
2021-11-05 - Learning from Comparisons - Stratis Ioannidis
2021-10-29 - Ordinal Patterns for Nonlinear Time Series Analysis - Thomas Stemler (starts at 10:00)
2021-10-22 - Satellite remote sensing (Part I) - Sivasakthy Selvakumaran (starts at 15:00)
2021-10-15 - No talk this week
2021-10-08 - Information Transfer for Engineering Fleets: Multi-Task Learning with Hierarchical Bayes - Lawrence Bull
2021-10-01 - Climate inference on daily rainfall across the Australian continent, 1876–2015 - Edward Cripps
2021-09-24 - No reading group this week
2021-09-17 - Statistical Finite Elements via Langevin Dynamics - Omer Deniz Akyildiz
2021-09-10 - Risk Based Structural Integrity Management - Domenic Di Francesco

Easter Term 2021

2021-07-23 - Hierarchical Statistical Models for Industrial Collaborative Prognosis - Maharshi Dhada
2021-07-16 - Adaptive Multilevel Delayed Acceptance - Tim Dodwell
2021-07-09 - Efficient stochastic optimisation by unadjusted Langevin Monte Carlo - Valentin De Bortoli
2021-07-02 - Variational inference for nonlinear ordinary differential equations - Sanmitra Ghosh
2021-06-25 - Neural (Stochastic) Differential Equations in Machine Learning - Patrick Kidger - Relevant papers: Neural ODEs, Neural CDEs, Neural SDEs
2021-06-18 - Transforming Gaussian processes with normalizing flows - Juan Maroñas
2021-06-11 - Solving high-dimensional Hamilton-Jacobi-Bellman PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space - Nik Nuesken
2021-06-04 - Gaussian Processes for Ordinal Regression - Ben Boys
2021-05-28 - Probabilistic modeling and identification of parameter fields with bounds: Application to solid mechanics - Hussein Rappel - Relevant work (starts at 15:00)
2021-05-21 - Theoretical Statistics - Andrius Ovsianas (starts at 15:00) (slides)
2021-05-14 - Nudging the particle filter - Omer Deniz Akyildiz
2021-05-07 - Non-stationary kernels for Gaussian processes - Kangrui Wang (starts at 15:30)
2021-04-30 - Continuous Time Limit of Gradient Algorithms - Justin
2021-04-23 - Statistical finite element methods for nonlinear PDEs - Connor

Lent Term 2021

In this term, we followed the “Applied Stochastic Differential Equations” book by Särkkä and Solin.

2021-04-16 - Ch 12 Stochastic Differential Equations in Machine Learning - Alessandro
2021-03-26 - Ch 11 Parameter Estimation in SDE Models - Alex
2021-03-19 - Ch 10 Filtering and Smoothing Theory - Connor
2021-03-12 - Ch 9 Approximation of Nonlinear SDEs - Hussein
2021-03-05 - Ch 8 Numerical Simulation of SDEs - Jan (notes, notebook)
2021-02-26 - An overview of nonasymptotic analysis for the stochastic gradient Markov chain Monte Carlo - Deniz
2021-02-19 - Ch 7 Useful Theorems and Formulas for SDEs - Justin (notes)
2021-02-12 - Ch 6 Statistics of Linear Stochastic Differential Equations - Ben (notes)
2021-01-29 - Ch 5 Probability Distributions and Statistics of SDEs - Yannis (notes)
2021-01-22 Ch 4 Itô Calculus and Stochastic Differential Equations (notes)

Michaelmas Term 2020

In this term, we focused on the topic of approximate Bayesian computation and Stein methods.

2020-12-11 Stochastic Variational Gradient Descent (notes)

Jan will present the Stochastic Variational Gradient Descent paper.

The main reference is: Liu, Q. and D. Wang (2016). Stein variational Gradient descent: a general purpose Bayesian inference algorithm. Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona, Spain, Curran Associates Inc.: 2378–2386. arXiv link

Other useful resources are:

2020-12-04 Score matching, Kernel Stein Discrepancy (notes)

Alessandro will present the following three papers:

  1. Barp, Alessandro, Francois-Xavier Briol, Andrew Duncan, Mark Girolami, and Lester Mackey. “Minimum Stein Discrepancy Estimators.” Paper presented at the Advances in Neural Information Processing Systems, 2019. NeurIPS link
  2. Dawid, Alexander Philip, and Monica Musio. “Theory and Applications of Proper Scoring Rules.” METRON 72, no. 2 (2014): 169-83. DOI link. research gate link.
  3. Hyvärinen, Aapo. “Estimation of Non-Normalized Statistical Models by Score Matching.” Journal of Machine Learning Research 6, no. Apr (2005): 695-709. JMLR link
2020-11-27 Introduction to Kernel Methods II with the focus on MMD

We will discuss the second part of the MLSS 2020 lecture on Kernel Methods by Arthur Gretton, with the focus on MMD.

2020-11-20 Introduction to Kernel Methods I (notes)

Justin will present Arthur Gretton’s introductory lecture into kernel methods.

2020-11-13 Sequential Monte Carlo without Likelihoods (notes)

Andrius will present the following paper:

Sisson, S. A., Y. Fan, and M. M. Tanaka. “Sequential Monte Carlo without Likelihoods.” Proceedings of the National Academy of Sciences 104, no. 6 (February 6, 2007): 1760–65. https://doi.org/10.1073/pnas.0607208104.

2020-11-06 ABC Methods for Climate Modelling (notes)

Yannis will present present chapter 19 of the Handbook of Approximate Bayesian Computation.

Sisson, S. A., Y. Fan, and M. A. Beaumont, eds. 2019. Handbook of Approximate Bayesian Computation. Boca Raton: CRC Press, Taylor and Francis Group.

2020-10-30 Summary Statistics (notes)

Alex will present chapter 5 of the Handbook of Approximate Bayesian Computation.

Sisson, S. A., Y. Fan, and M. A. Beaumont, eds. 2019. Handbook of Approximate Bayesian Computation. Boca Raton: CRC Press, Taylor and Francis Group.

2020-10-23 Statistical Inference for Noisy Nonlinear Ecological Dynamic Systems (notes)

Connor will present the following paper by Simon Wood.

Wood, Simon N. 2010. ‘Statistical Inference for Noisy Nonlinear Ecological Dynamic Systems’. Nature 466 (7310): 1102–4. https://doi.org/10.1038/nature09319.

2020-10-16 Introduction to Approximate Bayesian Computation (notes)

Ieva will present the introductory chapter of the Handbook of Approximate Bayesian Computation.

Sisson, S. A., Y. Fan, and M. A. Beaumont, eds. 2019. Handbook of Approximate Bayesian Computation. Boca Raton: CRC Press, Taylor and Francis Group.