........................................ Garo Panikian


Email: garo@cs.york.ac.uk

Hi !

I am a part-time PhD student in my third year of research in the Artificial Intelligence group at the University of York and I work under the supervision of Dr. James Cussens.

My research topic is about "Statistical Inference of Dynamical Systems with Application to Modelling Fish Population", which includes many aspects of Bayesian inference and time-series analysis.

Talks

Publications

 

Supporting Materials

Data

List of datasets analysed in "Maximum likelihood and Bayesian methods to identify and quantify heteroscedasticity in fisheries" is available here.

Density Plots

Density plots analysing the "Maximum likelihood and Bayesian methods to identify and quantify heteroscedasticity in fisheries" for illustrating the distribution of the coefficient of heteroscedasticity using frequentist (quasi-Newton) and Bayesian (Hamiltonian Monte Carlo) methods appied on the 257 fish populations.

Source Code

R code implementing the "Maximum likelihood and Bayesian methods to identify and quantify heteroscedasticity in fisheries".

This download includes everything required to reproduce results in the paper where the analysis is applied on 257 fish populations. It includes the datasets and a README.txt file within the download for instructions.

 

Prior to that, I completed my MSc degree from the University of Manchester, with distinction. My thesis was about Learning the correspondence between images and sounds of talking faces. The obtained generative model does not make use of labels, clustering or any other conventional technique. However, it is based on an unsupervised probabilistic learning. The following is an example of a synthesised animation, given speech, compared to Real Video.

A copy of my current Curriculum Vitae can be found here.


York Centre for Complex Systems Analysis, RCH 230-2, The Hub, Deramore Lane , University of York, York YO10 5GE, UK.