Complexity Explorer Challenge 2018 April

Problem Statement

Name	Purpose
201804.bib	Bibtex input to documentation
201804.xml	Workspace for Notepad++
challenge.nlogo	New model using links for assignment to pool
challenge.Rmd	Writeup for submission to SFI
get-data.r	Script to initialize data for challenge.Rmd
ieee-with-url	Required by challenge.Rmd to build bibliography
Proposal.tex	Agent Based Modelling Proposal
README.md	This File
sheep-wall-street.nlogo	Original Testbed for evaluating alternative strategies
strategies.csv	Parameters for strategies
View.jpg	Image for Info tab of take2.nlogo

Date	Remarks
24 Apr 2018	Started. Created repository and Wrote up Proposal
27 Apr 2018	Started implementing focal rules as in the original El Farol paper. Also started reading the Kolkaka Paise restaurant papers. Think about criteria for evaluating rules, e.g. review every n iterations to see whther I could do better, review against target performance, etc.
8 May 2018	Started on machine Learning after Fogel. I need to think through starting conditions. Is there enough time to start the way Fogel does?
9 May 2018	Started new version of model, using links for assignments to pools. Size of Fish (investors) indicates wealth, and happ/sad face indicates whether pool has paid off.
10 May 2018	Performed experiments on function closure. Bindings captured as doco indicates. Also function can recurse. So will use a function for predictor, withone argument indicating whther we want to initialize, predict or train.
11 May 2018	Learning is converging to same estimates
13 May 2018	I've digressed into working out how to use RMarkdown for writing up the challenge
23 May 2018	Digression into game theory
24 May 2018	Experiencers need to collect knowledge first. This is an optimal stopping problem