International Journal For Multidisciplinary Research

E-ISSN: 2582-2160   •   Impact Factor: 9.24

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 8, Issue 6 (November-December 2026) Submit your research before last 3 days of December to publish your research paper in the issue of November-December.

A Review Paper on Tiny ML and Deep Learning: Architectures, Optimization Techniques, Applications, and Future Directions

Author(s) Ms. Kirandeep Kaur, Prof. Birpal Singh, Prof. Guneet Kumar
Country India
Abstract Artificial Intelligence (AI) is growing rapidly, driving a strong demand for smart systems that work faster, more accurately, and with less energy. Deep Learning (DL) sits at the center of this progress because it learns complex patterns directly from massive datasets. However, standard deep learning models require heavy computing power, large amounts of memory, and continuous electricity, making them nearly impossible to run on tiny electronic devices. Tiny Machine Learning (Tiny ML) solves this problem by shrinking and adapting AI models so they can run directly on low-power hardware like microcontrollers, smartwatches, and Internet of Things (IoT) sensors.
This paper reviews how TinyML and deep learning work together, covering their core designs, optimization methods, and everyday uses. It breaks down key model-shrinking techniques—such as quantization, pruning, knowledge distillation, and lightweight neural networks—that cut down computing demands. It also looks at real-world uses in healthcare, farming, factory automation, environmental tracking, and smart transportation, while addressing major hurdles like tight memory limits, battery drain, accuracy trade-offs, and security risks. Ultimately, combining TinyML with lightweight deep learning makes AI far more affordable, energy-efficient, and capable of making instant decisions right where data is collected.
Keywords Keywords: TinyML, Deep Learning, Artificial Intelligence, Internet of Things, Edge Computing, Neural Networks, Model Compression, Embedded Systems.
Field Computer Applications
Published In Volume 8, Issue 6, November-December 2026
Published On 2026-11-14

Share this