Statistical Measures For Defining Curriculum Scoring Function
This work addresses the challenge of optimizing training efficiency and accuracy for image classification, though it appears incremental as it builds on existing curriculum learning concepts with new scoring functions.
The paper tackles the problem of improving neural network training by introducing a curriculum learning strategy that uses statistical measures like standard deviation and entropy to score data difficulty, showing performance improvements on real image classification tasks with convolutional and fully-connected networks across multiple datasets.
Curriculum learning is a training strategy that sorts the training examples by some measure of their difficulty and gradually exposes them to the learner to improve the network performance. Motivated by our insights from implicit curriculum ordering, we first introduce a simple curriculum learning strategy that uses statistical measures such as standard deviation and entropy values to score the difficulty of data points for real image classification tasks. We empirically show its improvements in performance with convolutional and fully-connected neural networks on multiple real image datasets. We also propose and study the performance of a dynamic curriculum learning algorithm. Our dynamic curriculum algorithm tries to reduce the distance between the network weight and an optimal weight at any training step by greedily sampling examples with gradients that are directed towards the optimal weight. Further, we use our algorithms to discuss why curriculum learning is helpful.