Enhancing Accuracy in Deep Learning Using Random Matrix Theory
This work addresses the efficiency and accuracy of deep-learning models, offering practical insights for practitioners, though it appears incremental as it builds on existing RMT applications.
The paper tackles the problem of reducing parameters in deep neural networks (DNNs) and CNNs using random matrix theory (RMT) for layer pruning, resulting in a drastic reduction of parameters without reducing accuracy and even increasing accuracy for fully connected DNNs while decreasing variance.
We explore the applications of random matrix theory (RMT) in the training of deep neural networks (DNNs), focusing on layer pruning that is reducing the number of DNN parameters (weights). Our numerical results show that this pruning leads to a drastic reduction of parameters while not reducing the accuracy of DNNs and CNNs. Moreover, pruning the fully connected DNNs actually increases the accuracy and decreases the variance for random initializations. Our numerics indicate that this enhancement in accuracy is due to the simplification of the loss landscape. We next provide rigorous mathematical underpinning of these numerical results by proving the RMT-based Pruning Theorem. Our results offer valuable insights into the practical application of RMT for the creation of more efficient and accurate deep-learning models.