Tight Genes: Intro to Genetic Algorithms
summary
Claims that genetic algorithms find “good enough” solutions fast backed by a runnable Ruby code examples knapsack and traveling‑salesman demo with populations (10→1000), mutation rate (≈1-in-4), and stopping rules (best unchanged for 100 gens). Mentions MATLAB and common libraries, shows fitness rising (20,000 → 26,000) and sensible final routes, and cites the real win: NASA ST5 antenna designed by a GA, so practical, reproducible results and concrete parameters are provided.
AI-written summary · about
chapters (15)
- 0:00Genetic algorithms intro
- 3:24How GA works step by step
- 6:37Knapsack initial candidates
- 11:17Fitness function & termination
- 15:55Selecting breeders & breeding
- 18:45Crossover & mutation in GA
- 23:49Running genetic algorithms repeatedly for best results
- 24:31Solving traveling salesman problem with genetic algorithms
- 30:16Roulette wheel selection
- 33:08Breeding & mutating TSP routes
- 37:46Other GA applications
- 38:49Recap: GA simplicity & versatility
- 41:01When not to use GA
- 43:12Self-mutating GA & LLMs
- 45:44Closing & Q&A
related talks
48mTight Genes: Intro to Genetic Algorithms
46mTight Genes: Intro to Genetic Algorithms
57mSolutions That Evolve: Building Self-Improving Systems with Genetic Algorithms
30mEvolutionary Algorithms Perfecting the Art of "Good Enough"
24mGenetic Algorithms in Go
35mThe "NumPy" Approach
46mBeing Creative with Genetic Algorithms and Typeclasses
42mLet's Talk About Sex: Breeding Programs with Evolution