Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video.
Jie WuGuanbin LiSi LiuLiang LinPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- temporal information
- tree structure
- video data
- markov decision process
- action selection
- natural language
- space time
- video sequences
- tree structured data
- video streams
- partially observable
- state and action spaces
- state space
- spatio temporal
- language learning
- markov decision problems
- model free
- video content
- reward function
- control policies
- partially observable environments
- action space
- decision problems
- video analysis
- actor critic
- real time
- video clips
- video retrieval
- video frames
- structured data
- xml files
- multimedia
- rl algorithms
- function approximation
- policy evaluation
- reinforcement learning problems
- spatio temporally
- average reward
- markov decision processes
- programming language
- state action
- function approximators
- partially observable markov decision processes
- temporal difference
- infinite horizon
- optimal control
- key frames
- multimedia data
- multi agent
- machine learning