Compute the Pareto front for the two-parameter, two-criteria example from linear-combination.ipynb more systematically. Plot or appropriately visualize both the Pareto front (in objective space) and the set of Pareto-optimal parameterizations (in parameter space).
The aim of this problem is to try and work with multicriteria optimization (MCO) in practice:
Any non-trivial decision problem where there are potentially conflicting objectives can be analysed as an MCO problem. Think of one from your domain of interest (broadly related to your presentation topic). Create a simple qualitative model of that problem and visualize the conflict between the objectives as a Pareto front.
The model in this case needs to be a multicriteria objective function y = f(x), mapping parameter values x to objectives y.
You probably have to simplify the problem and make many assumptions to come up with a model - that is OK! For the present purpose, the model does not need to be quantitatively accurate and account for everything. But it should reflect the main qualitative point, namely, that there are multiple objectives that can come to contradict one another, and that different choices need to be made depending on how much weight is given to each of these objectives. Therefore, do not simplify your model so far that the conflict between your objectives disappears.
If appropriate please continue to advise on the glossary; the concepts included this time were agent, optimization objective, optimization parameter, Pareto optimality, and rationality. Many thanks!
(submit through Canvas by end of 1st September 2023)