Fetching the paper…

Study on the Large Batch Size Training of Neural Networks Based on the Second Order Gradient · Around