Curriculum Learning with Deep Convolutional Neural Networks

By Vanya Avramova. 2015

23 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Jiashi Feng
2. Tat-seng Chua
3. Dacheng Tao
4. Yoshua Bengio
5. Yi Yang
6. Yudong Chen
7. Trevor Darrell
8. Eric Learned-Miller

Citations in the corpus

• Active Bias: Training More Accurate Neural Networks by Emphasizing High Variance Samples (2017, methodology): “As in SGD-WD, SGD-WE or self-paced learning [4], we train an unbiased model for several burnin epochs at the beginning so as to judge the sampling uncertainty reasonably and stably.”
• A Survey on Curriculum Learning (2021, background): “For example, Avramova [3] finds that convolutional neural networks derive most learning values from the hardest examples, and the damage of excluding those easiest examples is minor.”
• Deep Q-Networks for Accelerating the Training of Deep Neural Networks (2016, methodology): “Also based on the entropy of training samples, self-paced learning methods[8] are, however, usually coupled with the original optimization problems and, as far as we know, there is no work successfully applying self-paced learning to deep nets training [1].”