Novel Dynamic Batch-Sensitive Adam Optimiser for Vehicular Accident Injury Severity Prediction

TL;DR

This paper introduces DBS-Adam, a novel optimizer that dynamically adjusts learning rates based on batch difficulty, improving training stability and accuracy in imbalanced accident severity prediction tasks.

Contribution

The study presents DBS-Adam, an adaptive optimizer tailored for imbalanced sequential data, demonstrating superior performance over existing optimizers in accident severity classification.

Findings

DBS-Adam achieves 95.22% test accuracy.

It outperforms state-of-the-art optimizers with statistically significant precision improvements.

The framework enables effective real-time accident severity classification.

Abstract

The choice of optimiser is important in deep learning, as it strongly influences model efficiency and speed of convergence. However, many commonly used optimisers encounter difficulties when applied to imbalanced and sequential datasets, limiting their ability to capture patterns of minority classes. In this study, we propose Dynamic Batch-Sensitive Adam (DBS-Adam), an optimiser that dynamically scales the learning rate using a batch difficulty score derived from exponential moving averages of gradient norms and batch loss. DBS-Adam improves training stability and accelerates convergence by increasing updates for difficult batches and reducing them for easier ones. We evaluate DBS-Adam by integrating it with Bi-Directional LSTM networks for accident injury severity prediction, addressing class imbalance through SMOTE-ENN resampling and Focal Loss. Four experimental configurations compare baseline Bi-LSTM models and alternative architectures to assess optimiser impact. Rigorous comparison against state-of-the-art optimisers (AMSGrad, AdamW, AdaBound) across five random seeds demonstrated DBS-Adam's competitive performance with statistically significant precision improvements (p=0.020). Results indicate that DBS-Adam outperforms standard optimisation approaches, achieving 95.22% test accuracy, 96.11% precision, 95.28% recall, 95.39% F1-score, and a test loss of 0.0086. The proposed framework enables effective real-time accident severity classification for targeted emergency response and road safety interventions, demonstrating the value of DBS-Adam for learning from imbalanced sequential data.