- Try to decrease the value of the params.PopulationSize hyperparameter and see what happens. How did this affect the algorithm's performance?
- Try to set zero probabilities for the values of the following hyperparameters: params.ActivationFunction_SignedGauss_Prob, params.ActivationFunction_SignedSigmoid_Prob, and params.ActivationFunction_SignedSine_Prob. Was a successful solution found with these changes? How did this affect the configuration of the substrate connections?
- Print out the winning genome, try to come up with a visualization, then see how your intuition from looking at the genome matches with the visualized CPPN.
- Tech Categories
- Best Sellers
- New Releases
- Books
- Videos
- Audiobooks
Tech Categories Popular Audiobooks
- Articles
- Newsletters
- Free Learning
You're reading from Hands-On Neuroevolution with Python.
Iaroslav Omelianenko occupied the position of CTO and research director for more than a decade. He is an active member of the research community and has published several research papers at arXiv, ResearchGate, Preprints, and more. He started working with applied machine learning by developing autonomous agents for mobile games more than a decade ago. For the last 5 years, he has actively participated in research related to applying deep machine learning methods for authentication, personal traits recognition, cooperative robotics, synthetic intelligence, and more. He is an active software developer and creates open source neuroevolution algorithm implementations in the Go language.
Read more about Iaroslav Omelianenko
Unlock this book and the full library FREE for 7 days
Author (1)
Iaroslav Omelianenko occupied the position of CTO and research director for more than a decade. He is an active member of the research community and has published several research papers at arXiv, ResearchGate, Preprints, and more. He started working with applied machine learning by developing autonomous agents for mobile games more than a decade ago. For the last 5 years, he has actively participated in research related to applying deep machine learning methods for authentication, personal traits recognition, cooperative robotics, synthetic intelligence, and more. He is an active software developer and creates open source neuroevolution algorithm implementations in the Go language.
Read more about Iaroslav Omelianenko