DRL for Optimization
Thanh toán khi bàn giao
Build DRL algorithm eg Policy Gradient or DQN.
Build environment for scalability.
I provide specifications.
Two parts. Build 'environment' which is object class following template of gym. Environment with action space, statespace etc. Second part is the algorithm which runs the environment over and over determine optimal policy. An example pf such algo is Deep Q-learning
The Deep Q learning algorithm entails a neural network which is why pytorch is needed. Back to the environment: it model cashflows of an insurance policy and the underlying allocation of the premium to assets.
ID dự án: #36625519