[HTML][HTML] A review of green artificial intelligence: Towards a more sustainable future
Green artificial intelligence (AI) is more environmentally friendly and inclusive than
conventional AI, as it not only produces accurate results without increasing the …
conventional AI, as it not only produces accurate results without increasing the …
To talk or to work: Flexible communication compression for energy efficient federated learning over heterogeneous mobile edge devices
Recent advances in machine learning, wireless communication, and mobile hardware
technologies promisingly enable federated learning (FL) over massive mobile edge devices …
technologies promisingly enable federated learning (FL) over massive mobile edge devices …
Communication compression techniques in distributed deep learning: A survey
Nowadays, the training data and neural network models are getting increasingly large. The
training time of deep learning will become unbearably long on a single machine. To reduce …
training time of deep learning will become unbearably long on a single machine. To reduce …
[HTML][HTML] A survey: Distributed Machine Learning for 5G and beyond
Abstract 5 G is the fifth generation of cellular networks. It enables billions of connected
devices to gather and share information in real time; a key facilitator in Industrial Internet of …
devices to gather and share information in real time; a key facilitator in Industrial Internet of …
Bidirectional compression in heterogeneous settings for distributed or federated learning with partial participation: tight convergence guarantees
C Philippenko, A Dieuleveut - arXiv preprint arXiv:2006.14591, 2020 - arxiv.org
We introduce a framework-Artemis-to tackle the problem of learning in a distributed or
federated setting with communication constraints and device partial participation. Several …
federated setting with communication constraints and device partial participation. Several …
Service delay minimization for federated learning over mobile devices
Federated learning (FL) over mobile devices has fostered numerous intriguing
applications/services, many of which are delay-sensitive. In this paper, we propose a service …
applications/services, many of which are delay-sensitive. In this paper, we propose a service …
Improved step-size schedules for proximal noisy gradient methods
Noisy gradient algorithms have emerged as one of the most popular algorithms for
distributed optimization with massive data. Choosing proper step-size schedules is an …
distributed optimization with massive data. Choosing proper step-size schedules is an …
Over-the-air computation for distributed systems: Something old and something new
Facing the upcoming era of Internet-of-Things and connected intelligence, efficient
information processing, computation, and communication design becomes a key challenge …
information processing, computation, and communication design becomes a key challenge …
Green Federated Learning: A new era of Green Aware AI
The development of AI applications, especially in large-scale wireless networks, is growing
exponentially, alongside the size and complexity of the architectures used. Particularly …
exponentially, alongside the size and complexity of the architectures used. Particularly …
Adaptive compression for communication-efficient distributed training
We propose Adaptive Compressed Gradient Descent (AdaCGD)-a novel optimization
algorithm for communication-efficient training of supervised machine learning models with …
algorithm for communication-efficient training of supervised machine learning models with …