https://github.com/eeoooue/simulated-annealing
Interactive demo featuring Simulated Annealing (SA) as applied to the Travelling Salesman Problem (TSP).
https://github.com/eeoooue/simulated-annealing
metaheuristics simulated-annealing travelling-salesman-problem
Last synced: 29 days ago
JSON representation
Interactive demo featuring Simulated Annealing (SA) as applied to the Travelling Salesman Problem (TSP).
- Host: GitHub
- URL: https://github.com/eeoooue/simulated-annealing
- Owner: eeoooue
- Created: 2024-08-01T08:13:23.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2024-08-15T06:11:19.000Z (8 months ago)
- Last Synced: 2024-08-15T07:40:00.937Z (8 months ago)
- Topics: metaheuristics, simulated-annealing, travelling-salesman-problem
- Language: Jupyter Notebook
- Homepage: https://eeoooue.github.io/simulated-annealing/
- Size: 360 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Simulated Annealing
Simulated Annealing is a single-state method for combinatorial optimization.
Inspired by annealing in metallurgy, this metaheuristic strategy provides a dynamic temperature variable as a means of controlling the compromise between **exploration** and **exploitation** during the optimization process.
### Interactive Demo
While the cooling of the system is typically automated, you can try performing simulated annealing by hand using this [interactive demo](https://eeoooue.github.io/simulated-annealing/), which presents a typical _Travelling Salesman Problem_ for a map featuring 50 locations.
To emulate the annealing process, first 'heat up' the system by raising the temperature until it appears 'melted'.
Then, gradually 'cool' the system at a steady rate to hopefully arrive at a high quality solution.
### Possible Solutions
There are a staggering 608,281,864,034,267,560,872,252,163,321,295,376,887,552,831,379,210,240,000,000,000 possible solution states for the problem featured in this project.
**Monte Carlo Sampling** [has been used](./python/Monte-Carlo-Sampling.ipynb) to estimate the distribution of path lengths among these possible solutions.
