Probabilistic Contrastive Learning Recovers the Correct Aleatoric Uncertainty of Ambiguous Inputs.
Michael KirchhofEnkelejda KasneciSeong Joon OhPublished in: ICML (2023)
Keyphrases
- learning algorithm
- learning tasks
- reinforcement learning
- learning process
- uncertain data
- supervised learning
- online learning
- learning systems
- active learning
- knowledge acquisition
- inference process
- database
- learning analytics
- inductive learning
- learning scenarios
- learning activities
- prior knowledge
- knowledge base
- search engine
- information retrieval
- neural network