Planning sequential interventions to tackle depression in large uncertain social networks using deep reinforcement learning.
Aye Phyu Phyu AungSenthilnath JayaveluXiaoli LiBo AnPublished in: Neurocomputing (2022)
Keyphrases
- reinforcement learning
- social networks
- planning under partial observability
- action selection
- partial observability
- deterministic domains
- heuristic search
- planning problems
- online social networks
- partially observable
- state space
- social network analysis
- dynamic programming
- macro actions
- possibilistic networks
- ai planning
- network structure
- stochastic domains
- optimal policy
- markov decision problems
- blocks world
- complex domains
- decision support
- decision making
- social media
- learning algorithm
- temporal difference
- markov decision processes
- information diffusion
- model free
- single agent
- social network data
- community detection
- function approximation
- online communities
- domain independent
- transfer learning
- graphical models
- mobile robot