Next-Generation TinyML: Innovations in Models, Security, and Applications for Constrained Intelligent Systems

Dear Colleagues,

This Special Issue, "Next-Generation TinyML: Innovations in Models, Security, and Applications for Constrained Intelligent Systems", takes a close look at the most exciting recent steps forward in Tiny Machine Learning (TinyML).

TinyML is basically revolutionizing the way we put real artificial intelligence onto incredibly tiny, low-power devices—the kind that barely have any memory, battery, or processing power to spare.

We are especially interested in the kinds of innovations that actually make TinyML stronger, safer, and more usable in the real world. These breakthroughs let us run smart, energy-sipping AI straight on microcontrollers, edge sensors, and other embedded hardware, even when every kilobyte, milliwatt, and cycle counts.

This Special Issue also pays close attention to sustainability and privacy in edge computing—because we want the intelligence living on these little devices to be reliable, trustworthy, and genuinely helpful, no matter how harsh or remote the environment.

This Special Issue covers (but is not limited to) the following main pillars:

• Innovations in models—new ways to design compact architectures, smart compression tricks (quantization, pruning, knowledge distillation), hardware-aware tuning, plus fresh ideas like efficient transformers, tiny language models, and uncertainty-aware models that thrive under tight constraints.
• Security—solid ways to defend against adversarial attacks, side-channel leaks, and privacy risks—things like adapted federated learning, differential privacy, robust inference, uncertainty quantification to build confidence in decisions, and even blockchain/IOTA to provide tamper-proof logs, secure data origins, and decentralized trust in IoT networks.
• Applications—concrete, real-life examples that show real value: human behavior understanding, gesture and handwriting recognition (including culturally tailored versions), health wearables, driver monitoring systems, environmental and agricultural sensing, smart IoT setups, and predictive maintenance in industry.

We warmly welcome original research articles that present novel contributions and comprehensive reviews that offer valuable synthesis of the current state and future directions in this dynamic field.

Dr. Yassine Maleh
Prof. Dr. Khalid El Makkaoui
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Machine Learning and Knowledge Extraction is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• tiny machine learning
• edge intelligence
• model compression
• tiny language models
• resource-constrained AI
• adversarial robustness
• privacy-preserving machine learning
• on-device inference
• uncertainty quantification
• federated learning