Designing with Evolution

Free AI web copilot to create summaries, insights and extended knowledge, download it at here

1152

Abstract

he previous section, you can use evolution as a viable alternative to other machine learning approaches. Imagine you were a game designer tasked with designing new levels for a new puzzle game. Your task is to design 500 unique levels. Using a genetic algorithm, you could evolve levels from a collection of a few hand-designed levels to be sufficiently different from one another, adding in penalties for levels that are invalid or impossible to complete.Researchers<a href="https://readmedium.com/where-to-go-next-4231207bdb7d">[32]</a> in 2018 experimented and outlined an approach to evolutionary game design and developed a fitness function that rates evolved levels based on their playability. They also created recombination methods that enforced creativity between levels and ultimately proved that evolution is a viable approach to level design.Designing anything from games to antennas to websites is a practical application of genetic algorithms. As you’ve learned in this book, genetic algorithms are capable of intelligently searching through a large space of solutions to iteratively produce better and be

Options

tter solutions. One advantage of genetic algorithms in this space is they’re not bottlenecked by the limits of human design philosophy. It’s often difficult for human designers to break away from convention to come up with truly unique and revolutionary designs. Genetic algorithms don’t have these same limitations.👈 <a href="https://readmedium.com/learning-with-evolution-14543e8ec883">Learning with Evolution</a> | <a href="https://readmedium.com/table-of-contents-879fc8614df">TOC</a> | <a href="https://readmedium.com/trading-with-evolution-5d39bbebc853">Trading with Evolution</a> 👉Genetic Algorithms in Elixir by Sean Moriarity can be purchased in other book formats <a href="https://pragprog.com/titles/smgaelixir">directly from the Pragmatic Programmers</a>. If you notice a code error or formatting mistake, please let us know <a href="https://readmedium.com/how-to-report-errata-4e164674347a">here</a> so that we can fix it.<figure id="4bfa"><img src="https://cdn-images-1.readmedium.com/v2/resize:fit:800/1*U2E2a23SuM4520w9CUXSWw.jpeg"><figcaption></figcaption></figure></article></body>

Genetic Algorithms in Elixir — by Sean Moriarity (95 / 101)

NASA is responsible for perhaps the most famous application of genetic algorithms, as they used them to evolve the design of an antenna for maximum efficiency aboard their ST5[31] mission. They were able to produce a number of viable designs in a short amount of time — replacing the usual labor intensive work of designing by hand. Their algorithms also produced a number of unique designs that likely would never have been considered by human designers.

The concept of evolving designs extends to other fields as well. For example, game designers use artificial intelligence to develop unique levels. As you learned in the previous section, you can use evolution as a viable alternative to other machine learning approaches. Imagine you were a game designer tasked with designing new levels for a new puzzle game. Your task is to design 500 unique levels. Using a genetic algorithm, you could evolve levels from a collection of a few hand-designed levels to be sufficiently different from one another, adding in penalties for levels that are invalid or impossible to complete.

Researchers[32] in 2018 experimented and outlined an approach to evolutionary game design and developed a fitness function that rates evolved levels based on their playability. They also created recombination methods that enforced creativity between levels and ultimately proved that evolution is a viable approach to level design.

Designing anything from games to antennas to websites is a practical application of genetic algorithms. As you’ve learned in this book, genetic algorithms are capable of intelligently searching through a large space of solutions to iteratively produce better and better solutions. One advantage of genetic algorithms in this space is they’re not bottlenecked by the limits of human design philosophy. It’s often difficult for human designers to break away from convention to come up with truly unique and revolutionary designs. Genetic algorithms don’t have these same limitations.

Genetic Algorithms in Elixir by Sean Moriarity can be purchased in other book formats directly from the Pragmatic Programmers. If you notice a code error or formatting mistake, please let us know here so that we can fix it.