Zane SelvansTHE ELECTRIC SHEEP LIVE
GinoAppunti di ricerca sulla evoluzione di dinamiche cooperative in società simulate con algoritmi genetici
In una società virtuale, composta da agenti artificiali che interagiscono secondo le regole del dilemma del prigioniero con iterazione, emergono complesse dinamiche e strategie di relazione.
Lo studio si ispira ai lavori di Robert Axelrod, un politologo della Università del Michigan che per primo sperimentò tornei tra automi capaci di giocare al Prisoner's Dilemma. Consentendo agli agenti artificiali di modificare la tabella di payoff di partenza, Genagents rappresenta una evoluzione di quelle sperimentazioni.
La sintesi del lavoro qui pubblicata si compone di una introduzione alla logica del Dilemma del Prigioniero con iterazione, una descrizione del modello matematico del sistema di simulazione preceduta da una breve introduzione alle teorie evoluzionistiche che lo hanno ispirato. Sono infine indicati i principali risultati sperimentali ottenuti e alcune riflessioni sui possibili scenari interpretativi in relazione a dinamiche osservabili nel mondo reale.
AnthonyI came across a post on LinkedIn about evolutionary computation, and opted to post this in response:
I never stopped using evolutionary computation. I'm even weirder and use coevolutionary algorithms. Unlike EC, the latter have a bad reputation as being difficult to apply, but if you know what you're doing (e.g. by reading my publications 😉) they're quite powerful in certain application areas. I've successfully applied them to designing resilient physical systems, discovering novel game-playing strategies, and driving online tutoring systems, among other areas. They can inform more conventional multi-objective optimization.
Many challenging problems are not easily "vectorized" or "numericized", but might have straightforward representations in discrete data structures. Combinatorial optimization problems can fall under this umbrella. Techniques that work directly with those representations can be orders of magnitude faster/smaller/cheaper than techniques requiring another layer of representation (natural language for LLMs, vectors of real values for neural networks). Sure, given enough time and resources clever people can work out a good numerical re-representation that allows a deep neural network to solve a problem, or prompt engineer an LLM. But why whack at your problem with a hammer when you have a precision instrument?
I started to put up notes about (my way of conceiving) coevolutionary algorithms on my web site, here. I stopped because it's a ton of work and nobody reads these as far as I can tell. Sound off if you read anything there!
#AI #GenAI #GenerativeAI #LLMs #EvolutionaryComputation #GeneticAlgorithms #GeneticProgramming #EvolutionaryAlgorithms #CoevolutionaryAlgorithms #Cooptimization #CombinatorialOptimization #optimization
I never stopped using evolutionary computation. I'm even weirder and use coevolutionary algorithms. Unlike EC, the latter have a bad reputation as being difficult to apply, but if you know what you're doing (e.g. by reading my publications 😉) they're quite powerful in certain application areas. I've successfully applied them to designing resilient physical systems, discovering novel game-playing strategies, and driving online tutoring systems, among other areas. They can inform more conventional multi-objective optimization.
Many challenging problems are not easily "vectorized" or "numericized", but might have straightforward representations in discrete data structures. Combinatorial optimization problems can fall under this umbrella. Techniques that work directly with those representations can be orders of magnitude faster/smaller/cheaper than techniques requiring another layer of representation (natural language for LLMs, vectors of real values for neural networks). Sure, given enough time and resources clever people can work out a good numerical re-representation that allows a deep neural network to solve a problem, or prompt engineer an LLM. But why whack at your problem with a hammer when you have a precision instrument?
I started to put up notes about (my way of conceiving) coevolutionary algorithms on my web site, here. I stopped because it's a ton of work and nobody reads these as far as I can tell. Sound off if you read anything there!
#AI #GenAI #GenerativeAI #LLMs #EvolutionaryComputation #GeneticAlgorithms #GeneticProgramming #EvolutionaryAlgorithms #CoevolutionaryAlgorithms #Cooptimization #CombinatorialOptimization #optimization
AI Daily PostDiscover how TPOT uses genetic algorithms to evolve machine‑learning pipelines in just four steps—crossover, mutation, grid search and more—on the classic Iris dataset. A concise guide for Python enthusiasts who want automated model building. #TPOT #GeneticAlgorithms #MachineLearningPipelines #Python
🔗 https://aidailypost.com/news/tpot-evolves-ml-pipelines-via-genetic-algorithms-four-steps
Pragmatic Bookshelf 📚Genetic algorithms uncover solutions that brute force would miss, improving everything from shipping logistics to portfolio optimization.
Get Genetic Algorithms in Elixir by Sean Moriarity at https://pragprog.com/titles/smgaelixir/genetic-algorithms-in-elixir/
#elixir #geneticalgorithms #functionalprogramming
52°North🚀 As the Google Summer of Code 2025 comes to a close, our two students write about their work, challenges and solutions. Mayn thanks for your hard work!!
Check out their final blog posts: https://blog.52north.org/category/gsoc/
👉 #KomMonitor: Breathing New Life into an Open-Source Gem (Pranjal Goyal)
👉 Genetic Algorithm for Ship Route Optimization (Shreyas Ranganatha)
Chris Woody Woodruff🧬 Day 35, the final post of the Genetic Algorithms Bootcamp, is live!
Today: using GAs for creative art and design.
Evolution isn’t just for optimization. It can spark imagination, too.
Thanks to everyone who followed along, whether 1 post or all 35!
Day 34: Genetic Algorithms vs. Other Optimization Techniques: A Developer’s Perspective - Chris Woody Woodruff | Fractional Architect
Genetic Algorithms (GAs) are a powerful optimization strategy inspired by the principles of natural evolution. But they are far from the only technique in a developer's toolbox. In this post, we will compare Genetic Algorithms with other widely-used optimization methods such as Gradient Descent, Simulated Annealing, and Particle Swarm Optimization. The goal is to understand
🧬 Day 34 of the Genetic Algorithms Bootcamp is live!
Today, we compare GAs vs. other optimization techniques.
Where do GAs shine? Where do they fall short? A developer’s perspective.
Day 34: Genetic Algorithms vs. Other Optimization Techniques: A Developer’s Perspective - Chris Woody Woodruff | Fractional Architect
Genetic Algorithms (GAs) are a powerful optimization strategy inspired by the principles of natural evolution. But they are far from the only technique in a developer's toolbox. In this post, we will compare Genetic Algorithms with other widely-used optimization methods such as Gradient Descent, Simulated Annealing, and Particle Swarm Optimization. The goal is to understand
🧬 Day 33 of the Genetic Algorithms Bootcamp is live!
Case study: using GAs to optimize hyperparameters in a neural network.
Let evolution find better configs for smarter models.
Day 33: Case Study: Using a Genetic Algorithms to Optimize Hyperparameters in a Neural Network - Chris Woody Woodruff | Fractional Architect
Tuning hyperparameters for machine learning models like neural networks can be tedious and time-consuming. Traditional grid search or random search lacks efficiency in high-dimensional or non-linear search spaces. Genetic Algorithms (GAs) offer a compelling alternative by navigating the hyperparameter space with adaptive and evolutionary pressure. In this post, we’ll walk through using a Genetic Algorithm
🧬 Day 32 of the Genetic Algorithms Bootcamp is live!
Today, we’re tackling when GAs go wrong.
From poor performance to premature convergence, learn how to debug and keep evolution on track.
Day 32: When Genetic Algorithms Go Wrong: Debugging Poor Performance and Premature Convergence - Chris Woody Woodruff | Fractional Architect
Even well-written Genetic Algorithms can fail. You might see little improvement over generations, results clustering around poor solutions, or a complete stall in progress. These symptoms often point to premature convergence, loss of genetic diversity, or flaws in selection and fitness evaluation. Debugging GAs requires tools, insight, and techniques for diagnosis and correction. Understanding Premature