CAFE: Coarse-to-Fine Neural Symbolic Reasoning for Explainable Recommendation.
Yikun XianZuohui FuHandong ZhaoYingqiang GeXu ChenQiaoying HuangShijie GengZhou QinGerard de MeloS. MuthukrishnanYongfeng ZhangPublished in: CoRR (2020)
Keyphrases
- coarse to fine
- symbolic reasoning
- multiscale
- multiresolution
- symbolic representation
- production rules
- hierarchical segmentation
- image registration
- object detection
- neural network
- dynamic programming
- matching scheme
- hierarchical representation
- truth maintenance systems
- image pyramid
- biologically motivated
- knowledge based systems
- multi class
- knowledge representation
- pairwise
- reinforcement learning
- training data