Lazy vs hasty: linearization in deep networks impacts learning schedule based on example difficulty.
Thomas GeorgeGuillaume LajoieAristide BaratinPublished in: Trans. Mach. Learn. Res. (2022)
Keyphrases
- learning algorithm
- learning process
- connectionist networks
- incremental learning
- learning problems
- neural network
- inductive learning
- learning tasks
- unsupervised learning
- training data
- case study
- social networks
- multi agent systems
- active learning
- graphical models
- support vector
- genetic algorithm
- learning scheme
- deep learning
- database