activation function

3 papers

• Emergence and scaling laws in SGD learning of shallow neural networks
• Low Rank Gradients and Where to Find Them
• Minimum Width for Deep, Narrow MLP: A Diffeomorphism Approach