Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{185354,
        author = {Abhisri B, Assistant Professor in Department of Computer Science},
        title = {Unified Model for Deep Learning Optimization and Outlier Detection Using Hybrid Learning Techniques},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {5},
        pages = {987-993},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=185354},
        abstract = {In recent years, deep learning models have demonstrated remarkable success across a wide range of applications. However, their performance often deteriorates in the presence of noisy data and outliers. This research proposes a unified framework that combines advanced training optimisation techniques with integrated outlier detection to improve both the accuracy and robustness of deep neural networks. The model incorporates responsive activation functions, adaptive learning rate algorithms, dropout regularisation, and pretraining to optimise the learning process. Simultaneously, it employs autoencoders for unsupervised feature learning and integrates both clustering-based and classification-based methods for effective outlier identification. Experimental evaluations conducted on benchmark datasets demonstrate that the proposed hybrid approach enhances model generalization, reduces overfitting, and improves anomaly detection performance compared to traditional deep learning pipelines. The unified model shows potential for deployment in real-world scenarios where data quality and reliability are critical.},
        keywords = {},
        month = {November},
        }