Grad-Instructor: Universal Backpropagation with Explainable Evaluation Neural Networks for Meta-learning and AutoML

TL;DR

This paper introduces Grad-Instructor, a universal backpropagation method using explainable Evaluation Neural Networks trained via reinforcement learning to improve neural network training efficiency and accuracy in meta-learning and AutoML.

Contribution

It proposes a novel ENN-based evaluation approach integrated into backpropagation, enhancing training without increasing epochs and providing explainability through Grad-CAM.

Findings

Achieved 93.02% test accuracy on MLPs, 2.8% higher than conventional methods.

Reduced overfitting by balancing training and test errors.

Enabled efficient inference with input data processed at lower resolution.

Abstract

This paper presents a novel method for autonomously enhancing deep neural network training. My approach employs an Evaluation Neural Network (ENN) trained via deep reinforcement learning to predict the performance of the target network. The ENN then works as an additional evaluation function during backpropagation. Computational experiments with Multi-Layer Perceptrons (MLPs) demonstrate the method's effectiveness. By processing input data at 0.15^2 times its original resolution, the ENNs facilitated efficient inference. Results indicate that MLPs trained with the proposed method achieved a mean test accuracy of 93.02%, which is 2.8% higher than those trained solely with conventional backpropagation or with L1 regularization. The proposed method's test accuracy is comparable to networks initialized with He initialization while reducing the difference between test and training errors. These improvements are achieved without increasing the number of epochs, thus avoiding the risk of overfitting. Additionally, the proposed method dynamically adjusts gradient magnitudes according to the training stage. The optimal ENN for enhancing MLPs can be predicted, reducing the time spent exploring optimal training methodologies. The explainability of ENNs is also analyzed using Grad-CAM, demonstrating their ability to visualize evaluation bases and supporting the Strong Lottery Ticket hypothesis.