Making Darwin Proud: Coding Evolutionary Algorithms in Java
summary
Shows a working Java implementation that uses NASA Space Technology 5 antennas as a proof point: an evolutionary algorithm actually designed an antenna that flew in space. Implements Traveling Salesman with plain Java, uses NASA WorldWind for distances, an Angular/Bootstrap front end for live demos, and a cut-and-crossfill recombination plus swap mutation and replace-worst survivor selection. Population sizes, mutation rate and generation limits are tunable; demos converge in minutes and code is available for cloning.
AI-written summary · about
chapters (9)
- 0:00Introduction and warm-up
- 2:30Evolutionary algorithms
- 5:40Theory of evolution
- 11:01The Traveling Salesman problem
- 17:27Designing the algorithm
- 23:58Implementation details
- 36:35Algorithm tuning
- 38:17Live demo and results
- 45:42Frameworks and conclusion
related talks
48mFaith, Evolution, and Programming Languages: from Haskell to Java
01h 00mEvolutionary Design (StudentDay)
44mBehind the Scenes of Java's Evolution: The Case of Pattern Matching
28mJava for Education using SnapCode
20mJava's Hidden Gems: Tools and Libraries
46mLearning modern Java the playful way
52mLife After Dad Joke Apps Building “Enterprise Grade” AI Frameworks for Java
44mJava Evolution Vector: DOP
