Pseudo-label Learning with Calibrated Confidence Using an Energy-based Model.
Masahito TobaSeiichi UchidaHideaki HayashiPublished in: CoRR (2024)
Keyphrases
- prior knowledge
- computational model
- learning process
- learning scheme
- reinforcement learning
- probabilistic model
- learning problems
- mathematical model
- learning algorithm
- decision theoretic
- learning models
- formal model
- computational models
- learning tasks
- learning phase
- learned models
- em algorithm
- learning mechanism
- connectionist networks
- automatically learned
- conceptual model
- statistical model
- learning systems
- unsupervised learning
- markov random field
- decision trees