Anand

- [ ] Plot average performance vs iteration for all optimizers (#44) - [ ] Each optimizer should be run for multiple starting points and average fitness reported - [...

Implementation [available in scikit-learn](http://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html)

stop_criterion requires knowledge of optima

4

Which is not known in most cases. It should be defined as the delta change in fitness where the optimization is stopped.

To be consistent with the rest of the optimizers

Algorithms should handle discrete data

1

E.g. strings, or just a list of possible values.

Step size should be defined by the optimizee

1

We want the step size to be a hyper-parameter of the optimizee rather than an optimizer hyper-parameter, since each optimizee has different requirements, and may even have different required step...

Implement functions with saddle points in high dimensions

1

How do we construct such functions? (references)

Multi-dimensional fitness evaluation

1

Should it be a weighted sum (as it is currently) or a lexicographic ordering?

Ideas from David's work.