Cross Entropy in Deep Learning of Classifiers Is Unnecessary - ISBE Error Is All You Need.
Wladyslaw SkarbekPublished in: Entropy (2024)
Keyphrases
- deep learning
- cross entropy
- error function
- maximum likelihood
- unsupervised learning
- log likelihood
- decision trees
- machine learning
- language modeling
- training set
- training data
- mental models
- information retrieval
- learning rate
- evaluation metrics
- feature selection
- ranking functions
- weakly supervised
- support vector
- neural nets
- text classification
- activation function
- dimensionality reduction
- supervised learning