16 lines
503 B
Markdown
16 lines
503 B
Markdown
# Mathematical Foundations
|
|
|
|
This section provides formulas used across OptimizR algorithms.
|
|
|
|
## Differential Evolution
|
|
- Mutation and crossover follow classic DE/rand/1 and best/1 strategies.
|
|
- See Storn & Price (1997) for full derivations.
|
|
|
|
## MCMC
|
|
- Metropolis-Hastings with Gaussian proposals.
|
|
- Acceptance probability: $\alpha = \min\left(1, \frac{\pi(x')q(x\mid x')}{\pi(x)q(x'\mid x)}\right)$.
|
|
|
|
## HMM
|
|
- Baum-Welch (EM) for parameter estimation.
|
|
- Viterbi for decoding most likely state sequence.
|