https://github.com/filfreire/super-mario-experiments

metaheuristics experiments on super mario bros
https://github.com/filfreire/super-mario-experiments

Last synced: 4 days ago
JSON representation

metaheuristics experiments on super mario bros

• Host: GitHub
• URL: https://github.com/filfreire/super-mario-experiments
• Owner: filfreire
• License: mit
• Created: 2023-11-19T19:06:32.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2023-12-19T12:53:42.000Z (10 months ago)
• Last Synced: 2023-12-19T15:18:52.837Z (10 months ago)
• Language: Lua
• Size: 83 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# super-mario-experiments

metaheuristics experiments on super mario bros.

- Read the report [here](https://github.com/filfreire/super-mario-experiments/blob/main/report.pdf).

## Setup

> Note: tested using Windows 10 Pro 22H2

- Install [FCEUX](https://github.com/TASEmulators/fceux) (tested with win32 version)

- Load Rom (e.g. SMB NTSC USA rom)

- Try to run one of the `scripts` on fceux, alternatively run

  - `.\generate-initial-solution.ps1` for generating a feasible starting solution

  - `.\playback-solution.ps1 .\data\solutions1.txt` to try out a saved solution

  - `.\mario-genetic-1.ps1 ` to try out population based algorithm with a saved solution as the initial one

  - `.\mario-genetic-mutation-only.ps1 ` to try out custom evolutionary algorithm with a saved solution as the initial one

  - `.\generate-initial-solution-RW.ps1 `  for generating a feasible starting solution using random walk with a given random seed

  - `.\run-random-walk-experiment.ps1` to run a full random walk experiment of `.\generate-initial-solution-RW.ps1` be sure to edit the number of random seeds according to the number of threads of your CPU

### Extra setup steps

- Add `fceux` to the `PATH` in order to run it directly. For command line options see 

- Install `lua` (e.g. )

- Copy `iuplua51.dll` and `lfs.dll` from your Lua install directory into root of `fceux` installation directory

## Resources

- FCEUX lua scripting 

- Super Mario Bros RAM map 

- 

- 

- 

- 

- 

## Other extra sources

> Leane, M., & Noman, N. (2017, November). An evolutionary metaheuristic algorithm to optimise solutions to NES games. In 2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES) (pp. 19-24). IEEE.

>

> Aloupis, G., Demaine, E. D., Guo, A., & Viglietta, G. (2015). Classic Nintendo games are (computationally) hard. Theoretical Computer Science, 586, 135-160.

>

> Gabrielsen, C. (2012). Video Game Motion Planning reviewed NP-complete.

>

> Pelikan, M., & Goldberg, D. E. (2010). Genetic algorithms. MEDAL Report, (2010007), 1-28.

>

> Lobo, F. G., & Lima, C. F. (2007). Adaptive population sizing schemes in genetic algorithms. In Parameter setting in evolutionary algorithms (pp. 185-204). Berlin, Heidelberg: Springer Berlin Heidelberg.