Status Updates From Introduction to Evolutionar...

Nonstationary and Noisy Function Optimisation

An interesting chapter clearly illustrating the complexities of RL problems, and giving useful insight into different approaches to said challenges.

Hybridisation with Other Techniques: Memetic Algorithms

This as a particularly interesting chapter for me; I had previously heard of MAs but didn’t quite know what they were. Now that I have a better understanding of them I’m even tempted to explore them in practise at some point as I can now clearly visualise their usefulness.

Working with Evolutionary Algorithms

A very interesting chapter to end Part II. This brief chapter poses some very interesting questions, questions that are important to consider as you get started or be reminded of if you’ve worked in the field for a while. This time the one that really rang a bell is considering the purpose and the problem nature - design vs repetitive.

Parameter control

In conjunction with the previous chapter, these serve to clarify the difference between parameter tuning and parameter control, which might initially appear to be the same thing. It is an interesting chapter in itself presenting what work has been done so far, exploring pros and cons for each, and of combinations. I was particularly intrigued by the idea of varying populations, a novel idea to me.

Parameters and Parameter Tuning

A good overview with good tips in good practice. I found it very interesting that initially this aspect was so neglected, for me it currently seems like such an obvious thing to do. Very in accordance with the scientific method and stuff…

Popular Evolutionary Algorithm Variants

A very interesting chapter to end this first part of the book. It provides a comprehensive insight into some of the main evolutionary algorithms in use, giving the reader a general overview of how the ideas of evolution can be applied in different ways according to the problem at hand. Most interesting.

Fitness, selection and population management - as it often happens there are many existing techniques, yet often people tend to forget this and insist on using the same ol’ ones. This chapter provides a good overview of different techniques, making an emphasis of the importance of choosing carefully according to the problem, even considering the possibility of making combinations.

Clear overview of the importance of variance in GP alongside a comprehensive overview of the most common methods and how to best choose between them, including the possibility of combining. Good diagrams to illustrate the explanations.

Reading this... is gonna be tough. Interesting, but tough.