I´m new hear. Im a student of architecture and I´m working with grasshopper for three years now. I have written in my bachelor theses about digital morphogenesis. I`m very interested in topics like, evolutionary architecture, genetic algorithm, fitness landscape, galapagos, and learning how to code these things in python. My question is, do you guys have any tips for me, how to learn?
I prefer learning paths wich may take several months to a year, over doing short tutorials.
I´m not sure if I´m right, but is it true that machine learning, kind of includes my topic of interest?
That´s why I thought of bying these learning sources:
1.)machine learning for dummies, ISBN: 1119245516
2.)Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques for Building Intelligent Systems , ISBN: 978-1491962299
I´m open to any kind of suggestions.
Best wishes, Oliver
Did you read that from the Master David Rutten
Also, this thread:
A good source on these kinds of issues is Daniel Shiffman:
Genetic algorithms are a optimization method, they are useful when you need to explore a huge parametric space to find the configurations that best meet a criterion. Machine learning is something complete different, although it is also bio-inspired and can be used as a method of optimization, it is based on creating systems capable of learning and reasoning and are useful when you have a huge amount of data. Another optimization method that 20 years ago performed better than neural networks are the Support Vector Machines.
If you want to specialize in one or the other, I would recommend you 100% in ML. Cognitive computing is something relatively new (2010 I think, from deep learning) and there are not enough specialists to satisfy the possibilities of development at the moment. GA is not going to do anything new for architecture (I suppose), ML will do it.
As for morphogenesis, genetic algorithms are useful for discrete growth systems (such as DLA, L-Systems, cellular automata…), however for real morphogenesis (such as cellular systems or data-driven models), ML is more interesting to make them reactive or to analyze their emergence.
Thank you for the tip!
, but yes I did. It was one of sources for my bachelor thesis:)
Thanks a lot. There are some posts wich are new for me!
Thank you. that´s a very interesting input for me! I will do some google research on that. Some of the terms, I´m hearing the first time in my life. Maybe ML is a good starting point for me. I think I will do some further research, based on your recommondations before I decide wich course to choose:)
Best wishes, Oliver