EMERGE 2024

Enabling Machine Learning Operations for next-Gen Embedded Wireless Networked Devices
December 10, 2024

Call for Papers


The integration of Machine Learning (ML) and ML Operations (MLOps) onto Internet of Things (IoT) resource-constrained devices is revolutionising how smart wireless networks operate and evolve. ML techniques, evolving from self-supervised to reinforcement learning and beyond, offer rich solutions for optimising the efficiency and functionality of IoT networks. The focus on deploying and monitoring ML systems within these settings emphasises the need for innovative approaches that navigate the constraints of limited computational power, data availability, and energy resources. As MLOps methodologies evolve, they must address the dual challenges of integrating complex ML workflows into devices that are inherently limited in capacity while ensuring that these systems can be monitored, updated, and maintained with minimal overhead.

These technologies face the challenge of operating in environments with limited data, power, and computational resources, requiring models and pipelines that can adapt to new scenarios with minimal supervision and perform efficiently on low-resource devices. This demands a fusion of deep IoT system insights with advanced ML knowledge to develop practical solutions that enhance IoT networks' capabilities. The essence of deploying ML models in such contexts lies in leveraging advanced techniques that allow for real-time analytics, predictive maintenance, and autonomous decision-making on edge while operating within the stringent parameters of IoT devices.

This workshop wants to bring together researchers from the IoT and ML communities, to explore interdisciplinary approaches for developing ML algorithms and pipelines suited for IoT's unique environments where resources are a premium, and to engage in a lively debate on all facets of ML operations. The workshop will provide an avenue for learning about each other's challenges and methodologies and for debating future research agendas to jointly define the integration of ML on the IoT networks from a system perspective.

For the workshop's 1st edition, we invite researchers and practitioners from academia and industry to submit papers that focus on one of the following:

  • MLOps for resource-constrained IoT Systems
  • Privacy preservation within resource-constrained ML Pipelines
  • Useful tools and programming languages for deploying ML systems
  • Federated Learning and TinyML with on-board training
  • Deploying Continuous Learning strategies in ML systems
  • Continual learning, reinforcement learning
  • CI/CD pipelines for ML systems
  • Data challenges for deployed ML systems
  • Open problems in ML systems; challenges and blockers to the successful deployment of ML systems
  • Lightweight and real-time monitoring of ML systems
  • Self-Supervised Learning for Unsupervised IoT Data
  • Deploying hardware accelerators for ML pipelines in IoT systems
  • Benchmarking ultra-low power machine learning systems
  • Resource-efficient and adaptive deep learning

Important Dates


Paper submission [EXTENDED]: July 21, 2024 (AoE, 23:59 UTC-12)
Notification of acceptance: September 16, 2024
Camera-ready: October 6, 2024
Workshop date: December 10, 2024