HighMMT: Towards Modality and Task Generalization for High-Modality Representation Learning.
Paul Pu LiangYiwei LyuXiang FanShentong MoDani YogatamaLouis-Philippe MorencyRuslan SalakhutdinovPublished in: CoRR (2022)
Keyphrases
- learning process
- multi modal
- learning algorithm
- reinforcement learning
- wide range
- active learning
- online learning
- representational power
- data sets
- efficient learning
- concept learning
- image representation
- empirical studies
- probabilistic model
- multiscale
- face recognition
- multimedia
- artificial intelligence
- neural network