Answers are grounded in this paper and cite their sources.
Artificial Intelligence (AI) and Machine Learning (ML) are key technologies for creating intelligent systems in healthcare, speech processing, assistive technology, IoT, and other areas. Tiny Machine Learning (TinyML) enables ML models to run on microcontrollers with extremely limited computing, memory and power resources. In this research, we will look at intelligent edge systems and various challenges associated with them, such as computing constraints, model complexity and real-time requirements, as well as tradeoffs with energy, data security, and privacy. We also look at the challenges with cloud-based AI systems for healthcare and assistive technology. We will perform a literature review on TinyML and other embedded AI systems for constrained and resource-poor environments. We will look at different techniques to reduce the size of AI systems and to train them to run on resource-constrained systems. We will also look at other intelligent systems in speech, language and communication technologies for people with speech and language disabilities. We will look at different intelligent systems for healthcare and rehabilitation. Finally, we look at various technologies for the IoT, including intelligent systems with embedded intelligence for real-time analysis and decision making. We look at various types of intelligent systems and AI technologies, as well as the issues associated with integrating and deploying them in the real world. We look at barriers including reliability, security, privacy, and the tradeoffs with real-world deployment. We also look at various research approaches and evaluation criteria. We will propose plans of actions and research guidelines for developing more secure, explainable, and energy-efficient AI systems.
No takes yet. Share an insight, caveat, or question.
Mondal et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: