Real-Time Bidding with Soft Actor-Critic Reinforcement Learning in Display Advertising.
Daria YakovlevaArtem PopovAndrey FilchenkovPublished in: FRUCT (2019)
Keyphrases
- actor critic
- reinforcement learning
- real time
- display advertising
- temporal difference
- policy gradient
- reinforcement learning algorithms
- optimal control
- approximate dynamic programming
- function approximation
- neuro fuzzy
- online advertising
- gradient method
- control problems
- policy iteration
- state space
- control system
- average reward
- budget constraints
- reinforcement learning methods
- policy gradient methods
- action selection
- partially observable markov decision processes
- monte carlo
- model free
- convergence rate
- linear program
- heuristic search
- rl algorithms
- optimal policy
- supervised learning
- least squares
- multi agent
- natural actor critic
- machine learning