LogosQ Optimizer

Classical optimization algorithms for variational quantum algorithms, providing stable and fast parameter optimization.

Features

• Adam: Adaptive moment estimation with momentum
• L-BFGS: Quasi-Newton method for smooth objectives
• SPSA: Gradient-free stochastic approximation
• Natural Gradient: Fisher information-aware optimization
• GPU acceleration: Optional CUDA support

Quick Start

use logosq_optimizer::{Adam, Optimizer};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let optimizer = Adam::new()
        .with_learning_rate(0.01)
        .with_beta1(0.9);
    
    let mut params = vec![0.1; 16];
    let gradients = vec![0.01; 16];
    
    optimizer.step(&mut params, &gradients, 0)?;
    Ok(())
}

Installation

[dependencies]
logosq-optimizer = "0.1"

License

MIT OR Apache-2.0