Random walks¶
This is an archive of explorations, in the convex hull of Inference, Probability, Optimisation and Physics. I am putting this together to keep track of the things I study.
About¶
My name is Stratis. I am a second-year PhD student at the Computational and Biological Learning Lab (CBL) at the University of Cambridge. I am supervised by Carl Rasmussen and advised by Richard Turner. I have broad interests in probabilistic machine learning, see Google Scholar for my recent work.
New on Random Walks¶
[02/02/2023] Notes and demo on Annealed Importance Sampling.
[04/04/2022] Notes and demo on The Variational Gaussian Approximation Revisited.
[15/02/2022] Notes and demo on Global Inducing Point VI for BNNs.
[03/09/2021] Notes on the Gumbel distribution.
[01/09/2021] My first year report for the PhD at Cambridge.
[07/08/2021] Convex Optimisation notes on first three chapters of Boyd and Vandenberghe.
[09/06/2021] Notes and demo on Efficient Sampling from GP posteriors.
[05/05/2021] Notes and demo on Variational Sparse Gaussian Processes.
[30/04/2021] Notes and demo on Additive Gaussian Processes.
[19/04/2021] Notes and demo on Stein Variational Gradient Descent.
[02/04/2021] Notes and demo on Random Fourier Features.
[20/03/2021] Notes and demo on Interacting particle solutions of FPK equations through GLD estimation.
[10/03/2021] Short notes on the paper: Estimation of non-normalized statistical models by score matching.
[13/02/2021] Notes and a demo for the paper: Gaussian process approximations of stochastic differential equations.
[01/02/2021] Additional material on Conjugate Gradients.
[21/01/2021] Notes and demo on Natural Cubic Splinces.
[14/01/2021] Notes and demo on Adaptive Rejection Sampling.
[12/10/2020] Short notes on Conjugate Gradients
[10/09/2020] Short notes on the last three chapters of Grimmett and Welsh, An Introduction to Probability.
[06/09/2020] Short notes on the first nine chapters (out of 12) of Grimmett and Welsh, An Introduction to Probability.
[25/07/2020] A reproduction of the paper: Higham, An algorithmic Introduction to Numerical Simulation of SDEs.