1. Markov decision processes (MDPs) in R - Create and optimize MDPs with discrete time steps and state space. Both normal MDPs and hierarchical MDPs can be considered.
2. subplex - subplex solves unconstrained optimization problems using a simplex method on subspaces. The method is well suited for optimizing objective functions that are noisy or are discontinuous at the solution.
The subplex algorithm is due to Tom Rowan, ORNL.
Development Status: 5 - Production/Stable [Filter]
Intended Audience: Developers [Filter] , End Users/Desktop [Filter]
License: GNU General Public License (GPL) [Filter]
3. UCMINF - unconstrained optimization - UCMINF - An algoriithm for unconstrained non-linear optimization
The algorithm is of quasi-Newton type with BFGS updating of the inverse Hessian and soft line search with a trust region type monitoring of the input to the line search algorithm.
4. Optimization and solving packages - Packages for optimization, including large scale optimization and equation solving. Related numerical methods packages for gradient and hessian calculation and for factor rotation are also included.
5. Optimal Two-stage Phase II Design - This R package OptimPhase2 creates the optimal two-stage phase II designs with long-term endpoint. Optimality is defined to minimze either the expected sample size, expected duration of accrual and expected total study length.
Development Status: 5 - Production/Stable [Filter]