Centralized sub-critic based hierarchical-structured reinforcement learning for temporal sentence grounding.
Yingyuan ZhaoZhiyi TanBing-Kun BaoZhengzheng TuPublished in: Multim. Syst. (2023)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- temporal difference
- actor critic
- max margin learning
- natural language
- spatial and temporal
- spatio temporal
- temporal information
- learning algorithm
- policy gradient
- temporal constraints
- structured data
- temporal reasoning
- machine learning
- space time
- temporal patterns
- peer to peer
- temporal relations
- markov decision processes
- distributed environment
- transfer learning
- hierarchical reinforcement learning
- dynamic programming
- temporal databases
- state space
- information retrieval
- approximate dynamic programming
- optimal policy
- action selection
- text summarization
- temporal data
- neuro fuzzy