Computational Mechanics of Food Webs

Overview

This project is about estimating causal states from population time series data and reconstruction of an $\epsilon$-machine. One interesting question would be to look ant the causal states and find some correlations to nodes in a dynamic food web. Another question would be to quantify the population dynamics.

We have have a frequentist [1] and a Bayesian [2] reconstruction algorithm. It would be interesting to see if they come up with the same result. As input data we can use a food web simulation we have seen in Neos lecture, wich is available [here].

Reading:

[1] C. R. Shalizi, K. L. Shalizi, and J. P. Crutchfield. An algorithm for pattern discovery in time series. 2002.

[2] C. C. Strelioff, J. P. Crutchfield, and A. Hubler. Inferring markov chains: Bayesian estimation, model comparison, entropy rate, and out-of-class modeling. 2007.

Who's interested

• Olaf Bochmann
• ...
• ...

(yes, you can still add your name!)