🐕 Canine Olfactory Optimizer (COO)

A Novel Bio-Inspired Metaheuristic Algorithm
Inspired by how dogs track scents to find food

ðŸŽŊ Biological Inspiration: How Dogs Find Food

Watch how a pack of dogs uses their incredible sense of smell to locate food. Dogs employ multiple strategies: following scent gradients, casting about in zigzag patterns when they lose the trail, and cooperating with their pack to cover more ground.

ðŸķ
Dogs (Searching)
ðŸĶī
Dogs (Found Target)
🍖
Food Target
Scent Gradient (Stronger = Darker)
Dogs Searching
8
Average Distance
0
Dogs Found Food
0

⚙ïļ COO Algorithm: Interactive Hyperparameter Control

Explore how different hyperparameters control the behavior of the optimization algorithm. Adjust pack sizes, movement parameters, and cooperation strategies to see their effects in real-time.

Multiple packs explore different regions simultaneously
Number of dogs (agents) in each pack
Inertia - how much dogs maintain their current direction
Attraction to pack-local best position
Attraction to global best position (increases over time)
Sniffing noise - random local exploration
Casting behavior when dogs lose the trail
Magnitude of reacquisition movements
How often elite dogs are shared between packs
Percentage of top dogs using gradient-based search
Enable local gradient-based optimization
Enable information sharing between packs
Iteration
0
Best Fitness
0.00
Evaluations
0
Convergence Rate
0%

🏔ïļ Rosenbrock Function Optimization

The Rosenbrock function is a famous test case for optimization algorithms, featuring a long, narrow, parabolic valley. The global minimum is at (1, 1) with f(x) = 0. Watch how COO navigates this challenging landscape!

Rosenbrock Function: f(x, y) = (1 - x)Âē + 100(y - xÂē)Âē
Global Minimum: (x, y) = (1, 1), f(1, 1) = 0
Best Position
(0, 0)
Best Value
∞
Distance to Optimum
-
Success Rate
-