Align before Fuse: Vision and Language Representation Learning with Momentum Distillation.
Junnan LiRamprasaath R. SelvarajuAkhilesh Deepak GotmareShafiq R. JotyCaiming XiongSteven C. H. HoiPublished in: CoRR (2021)
Keyphrases
- learning algorithm
- learning process
- unsupervised learning
- real time
- knowledge acquisition
- feature hierarchies
- learning tasks
- prior knowledge
- multiscale
- computer vision
- active learning
- supervised learning
- bayesian networks
- image processing
- learning systems
- learning problems
- language learning
- inductive learning
- learning rate
- object oriented programming
- qualitative models
- data sets