Trust Region Policy Optimization of POMDPs.
Kamyar AzizzadenesheliManish Kumar BeraAnimashree AnandkumarPublished in: CoRR (2018)
Keyphrases
- trust region
- optimization methods
- line search
- partially observable markov decision processes
- optimal policy
- global optimum
- optimization algorithm
- optimization method
- hessian matrix
- optimization problems
- step size
- newton method
- column generation
- quadratic programming
- global convergence
- risk minimization
- reinforcement learning
- learning algorithm
- markov decision processes
- constrained optimization
- levenberg marquardt
- dynamical systems
- multi objective
- evolutionary algorithm
- search algorithm