Reinforcement Learning of Cooperative Persuasive Dialogue Policies using Framing.
Takuya HiraokaGraham NeubigSakriani SaktiTomoki TodaSatoshi NakamuraPublished in: COLING (2014)
Keyphrases
- cooperative
- reinforcement learning
- optimal policy
- multiagent reinforcement learning
- policy search
- control policies
- markov decision process
- multi agent
- reward function
- partially observable markov decision processes
- function approximation
- control policy
- state space
- markov decision processes
- multi agent reinforcement learning
- reinforcement learning agents
- fitted q iteration
- dialogue system
- dialogue management
- markov decision problems
- policy gradient methods
- reinforcement learning algorithms
- model free
- hierarchical reinforcement learning
- decision problems
- human machine
- multi agent systems
- decentralized control
- cooperative learning
- mixed initiative
- spoken dialogue systems
- machine learning
- conversational agent
- action space
- finite state
- human computer
- temporal difference
- learning algorithm
- natural language dialogue
- natural language
- long run
- neural network
- dynamic programming
- speech acts
- spoken language
- multiagent systems
- average cost
- learning process
- game theory
- infinite horizon
- policy iteration
- optimal control