Divide and Conquer: Leveraging Intermediate Feature Representations for Quantized Training of Neural Networks

Open in new window