A trajectory is worth three sentences: multimodal transformer for offline reinforcement learning.
Yiqi WangMengdi XuLaixi ShiYuejie ChiPublished in: UAI (2023)
Keyphrases
- reinforcement learning
- function approximation
- natural language
- multi modal
- real time
- fuzzy logic
- temporal difference
- fault diagnosis
- neural network
- multi document summarization
- multimodal interaction
- optimal policy
- state space
- learning process
- learning problems
- multimedia
- model free
- learning algorithm
- trajectory data
- trajectories of moving objects
- power transformers
- extractive summarization
- distribution network
- power system
- multi agent
- markov decision processes
- supervised learning