Regularizing Neural Networks by Penalizing Confident Output Distributions
This provides a simple and effective regularization technique for various machine learning tasks, though it is incremental as it builds on existing entropy-based methods.
The paper tackles the problem of overfitting in neural networks by penalizing low entropy output distributions, showing that this confidence penalty and label smoothing improve state-of-the-art models across six benchmarks, including image classification and machine translation, without hyperparameter tuning.
We systematically explore regularizing neural networks by penalizing low entropy output distributions. We show that penalizing low entropy output distributions, which has been shown to improve exploration in reinforcement learning, acts as a strong regularizer in supervised learning. Furthermore, we connect a maximum entropy based confidence penalty to label smoothing through the direction of the KL divergence. We exhaustively evaluate the proposed confidence penalty and label smoothing on 6 common benchmarks: image classification (MNIST and Cifar-10), language modeling (Penn Treebank), machine translation (WMT'14 English-to-German), and speech recognition (TIMIT and WSJ). We find that both label smoothing and the confidence penalty improve state-of-the-art models across benchmarks without modifying existing hyperparameters, suggesting the wide applicability of these regularizers.