31 lines
1.3 KiB
Markdown
31 lines
1.3 KiB
Markdown
# it3708 - bio-inspired artificial intelligence
|
|
|
|
## assignment 1
|
|
|
|
this project is written in [odin](https://odin-lang.org/) and [uiua](https://www.uiua.org/) ([uiua-plot](https://github.com/Omnikar/uiua-plot)) with [nix](https://nixos.org/) for dependency management.
|
|
|
|
with nix installed, just run
|
|
```
|
|
$ nix run
|
|
```
|
|
and it will run through the pipeline. it will generate data/image files in `output/`.
|
|
|
|
### implementation overview
|
|
|
|
i implement a genetic algorithm to solve the binary knapsack problem.
|
|
|
|
this is done by
|
|
- first generating a population of individuals/chromosomes, modeled as bit_arrays for memory performance.
|
|
- then continue for some fixed amount of generations
|
|
- calculate the fitness of each chromosome in the population by calculating the distance to the capacity. to avoid negative fitness values, chromosomes that overshoot the capacity are penalized.
|
|
- create offspring by
|
|
- selecting two parents using a tournament selection
|
|
- then performing single point crossover between the parents to create two children
|
|
- then mutate each child with random bit-flip mutations.
|
|
- create a new population via elitism survivor selection
|
|
- repeat
|
|
|
|
each generation, the fitness value maximum, minimum and mean values are logged in `output/data.csv`.
|
|
|
|
this data is read through uiua `src/plot.ua` to plot the data to `output/plot.png`.
|