Divide and Contrast: Self-supervised Learning from Uncurated Data.
Yonglong TianOlivier J. HénaffAäron van den OordPublished in: CoRR (2021)
Keyphrases
- data sets
- training data
- prior knowledge
- active learning
- raw data
- experimental data
- learning systems
- synthetic data
- statistical analysis
- data processing
- data sources
- hidden variables
- background knowledge
- image data
- high quality
- learning algorithm
- probabilistic model
- input data
- online learning
- data model
- computer systems
- data structure
- high dimensional data
- database systems
- machine learning
- learning tasks
- data mining
- complex data
- learning models
- database