Differential privacy techniques can help prevent overfitting and improve generalization in deep neural networks.
What is the best multi-stage architecture for object recognition? InComputer Vision, 2009 IEEE 12th International Conference on, pages 2146–2153
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Preventing overfitting in deep learning using differential privacy
Differential privacy techniques can help prevent overfitting and improve generalization in deep neural networks.