Reward Finetuning for Faster and More Accurate Unsupervised Object Discovery.
Katie LuoZhenzhen LiuXiangyu ChenYurong YouSagie BenaimCheng Perng PhooMark E. CampbellWen SunBharath HariharanKilian Q. WeinbergerPublished in: NeurIPS (2023)
Keyphrases
- supervised learning
- high quality
- reinforcement learning
- unsupervised learning
- highly accurate
- data driven
- data mining
- unsupervised manner
- long run
- high accuracy
- completely unsupervised
- databases
- computationally efficient
- semi supervised
- dynamic programming
- artificial neural networks
- data structure
- clustering algorithm
- website
- decision making
- feature selection