Continuous-Action Reinforcement Learning for Portfolio Allocation of a Life Insurance Company.
Carlo AbrateAlessio AngiusGianmarco De Francisci MoralesStefano CozziniFrancesca IadanzaLaura Li PumaSimone PavanelliAlan PerottiStefano PignataroSilvia RonchiadinPublished in: ECML/PKDD (4) (2021)
Keyphrases
- continuous action
- policy search
- reinforcement learning
- continuous state
- action space
- continuous state and action spaces
- partially observable markov decision processes
- reinforcement learning algorithms
- state space
- function approximation
- finite state
- robot navigation
- model free
- risk management
- markov decision processes
- decision making
- temporal difference
- optimal policy
- control policies
- learning algorithm
- dynamic programming
- policy gradient
- supervised learning
- reinforcement learning methods
- state action
- heuristic search
- function approximators
- state dependent
- markov decision problems
- markov chain
- transfer learning
- machine learning
- real valued