This function provides a unified interface to multiple algorithm packages, allowing different optimization algorithms to be selected for estimating optimal model parameters. The entire optimization framework is based on the log-likelihood returned by the model (or object function), making this function a collection of likelihood-based inference (LBI) methods. By abstracting over algorithm-specific implementations, the function enables flexible and consistent parameter estimation across different optimization backends.
Usage
estimate_1_LBI(env, model, lower, upper, control = list(), ...)Arguments
- env
multiRL.env
- model
Reinforcement Learning Model
- lower
Lower bound of free parameters
- upper
Upper bound of free parameters
- control
Settings manage various aspects of the iterative process, see control
- ...
Additional arguments passed to internal functions.