Chris Williams: Software

• Code by Gerrit van den Burg for the paper On Memorization in Probabilistic Deep Generative Models (van den Burg and Williams, NeurIPS 2021).
• Code by Michael Camilleri for the paper The Extended Dawid-Skene Model: Fusing Information from Multiple Data Schemas (Camilleri and Williams, ECML PKDD 2019 Workshops, CCIS 1167, pp. 121 - 136, 2020)
• Code by Taha Ceritli for the paper ptype: Probabilistic Type Inference (Ceritli, Williams and Geddes, Data Mining and Knowledge Discovery, 2020). Update Nov 2021: Code is now available for ptype-cat, an extension to ptype that can handle non-Boolean categorical variables. Paper: ptype-cat: Inferring the Type and Values of Categorical Variables pdf, Taha Ceritli and Christopher K. I. Williams. Presented at the ECML-PKDD Workshop on Automating Data Science, 17 Sept 2021
• Code by Pol Moreno for the paper Overcoming Occlusion with Inverse Graphics (Moreno, Williams, Nash, Kohli, 2016), including an updated version of Matt Loper's OpenDR.
• Code for Factorial Switching Linear Dynamical System (FSLDS) Monitoring of Intensive Care Unit Data. Chris, Williams, Partha Lal, Martin Shaw, September 2015. See accompanying technical report for more details.
• Matlab code for the paper Localisation microscopy with quantum dots using non-negative matrix factorisation, O. Mandula, I. Sumanovac Sestak, R. Heintzmann, C. K. I. Williams Optics Express, Vol. 22, Issue 20, pp. 24594-24605 (2014) download, or clone from github.
• Software developed for the paper A Framework for Evaluating Approximation Methods for Gaussian Process Regression, K. Chalupka, C. K. I. Williams and I. Murray, submitted for publication, 2012.
• Matlab code for the paper Multi-task Gaussian Process Prediction by Edwin V. Bonilla, Kian Ming A. Chai, and Christopher K. I. Williams (NIPS 2008).
• Matlab code for the paper Greedy Learning of Binary Latent Trees by S. Harmeling and C. K. I. Williams (In IEEE PAMI 33(6) 1087-1097, 2011).
• Matlab code giving a demonstration of a condition monitoring system for infants in intensive care developed by John Quinn in his PhD thesis.
• Software for object category recognition using the Generative Template of Features. For further details see Sprite Learning and Object Category Recognition using Invariant Features, PhD thesis by Moray Allan, School of Informatics, University of Edinburgh, 2007.
• GPML software to accompany the book Gaussian Processes for Machine Learning, C. E. Rasmussen and C. K. I. Williams, MIT Press, 2006.
• MATLAB code for 4th year project by Fabian Wauthier to implement and extend the ideas in Learning Patterns of Activity Using Real-time Tracking by C. Stauffer and W. E. Grimson (PAMI, 2000).
• MATLAB Toolbox for learning layers from video, see reference: Fast Unsupervised Greedy Learning of Multiple Objects and Parts from Video, M. K. Titsias and C. K. I. Williams, Proc. Generative-Model Based Vision Workshop, July 2004.
• Software developed for the paper Image Modelling with Position-Encoding Dynamic Trees, Amos J. Storkey, Christopher K. I. Williams, IEEE Trans Pattern Analysis and Machine Intelligence 25(7) 859-871 (2003)
• Software developed for the paper Combining belief networks and neural networks for scene segmentation, Xiaojuan Feng, C. K. I. Williams and S. N. Felderhof, IEEE Trans Pattern Analysis and Machine Intelligence 24(4) 467-483 (2002)
• Code is available to implement multi-class classification as described in Bayesian Classification with Gaussian Processes, C. K. I. Williams and David Barber, IEEE Trans Pattern Analysis and Machine Intelligence , 20(12) 1342-1351, (1998).
• GTM. Code in the netlab toolbox written by Ian Nabney implements the GTM model described in GTM: The Generative Topographic Mapping, C. M. Bishop, M. Svensen, C. K. I. Williams, Neural Computation 10(1) 215-234 (1998).
  
• An implementation of Using generative models for handwritten digit recognition, Revow, M. and Williams, C. K. I. and Hinton, G. E., IEEE Transactions on Pattern Analysis and Machine Intelligence 18(6) pp 592-606 (1996) is available from Michael Revow's homepage