Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection.
Thang DoanXin LiSima BehpourWenbin HeLiang GouLiu RenPublished in: CoRR (2023)
Keyphrases
- structure learning
- open world
- bayesian networks
- graphical models
- closed world
- incomplete information
- conditional independence
- transfer learning
- parameter estimation
- network structure
- knowledge base
- data exchange
- markov networks
- probability distribution
- belief propagation
- query answering
- maximum likelihood estimation
- machine learning
- sample size
- probabilistic model
- pairwise