Jensen-Shannon Divergence and Cross-Entropy Loss
Explore how PyTorch's JsdCrossEntropy loss function integrates Jensen-Shannon divergence with cross-entropy to measure differences in probability distributions. Understand its parameters and role in enhancing model learnability and robustness in image classification tasks.
We'll cover the following...
We'll cover the following...
The PyTorch Image Model provides a special loss called JsdCrossEntropy. As the name suggests, it uses Jensen-Shannon divergence and cross-entropy to calculate the total loss.
Jensen-Shannon divergence
The Jensen-Shannon divergence is a well-known term in the probability theory and statistics domains. It’s symmetric, has finite values, and measures ...