Nth Absolute Root Mean Error
This addresses the time vs. data size dilemma in neural network training for regression problems, offering a faster alternative to GPU-based solutions, though it appears incremental as a new loss function for a specific subset.
The paper tackles the slow training of neural networks on large regression datasets by introducing the Nth Absolute Root Mean Error (NARME) loss function, which reduces the required number of epochs to almost one-tenth compared to other loss functions while maintaining accuracy.
Neural network training process takes long time when the size of training data is huge, without the large set of training values the neural network is unable to learn features. This dilemma between time and size of data is often solved using fast GPUs, but we present a better solution for a subset of those problems. To reduce the time for training a regression model using neural network we introduce a loss function called Nth Absolute Root Mean Error (NARME). It helps to train regression models much faster compared to other existing loss functions. Experiments show that in most use cases NARME reduces the required number of epochs to almost one-tenth of that required by other commonly used loss functions, and also achieves great accuracy in the small amount of time in which it was trained.