Depthwise Convolution is All You Need for Learning Multiple Visual Domains.
Yunhui GuoYandong LiRogério Schmidt FerisLiqiang WangTajana RosingPublished in: CoRR (2019)
Keyphrases
- learning tasks
- learning process
- multiple tasks
- learning problems
- reinforcement learning
- learning systems
- inter related
- inductive inference
- incremental learning
- online learning
- low level
- learning algorithm
- neural network
- state space
- genetic algorithm
- elementary school
- learning scheme
- cross domain
- learning analytics
- e learning
- image processing
- support vector
- visual information
- transfer learning
- application domains
- prior knowledge
- visual features