PhyMask: An Adaptive Masking Paradigm for Efficient Self-Supervised Learning in IoT
This paper introduces PhyMask, an adaptive masking paradigm designed to enhance the
efficiency and interpretability of Masked Autoencoders (MAEs) in analyzing IoT sensing …
efficiency and interpretability of Masked Autoencoders (MAEs) in analyzing IoT sensing …
The Case for Micro Foundation Models to Support Robust Edge Intelligence
This paper advocates the concept of micro foundation models (μFMs), recently introduced by
the authors to describe a category of self-supervised pre-training solutions that we argue are …
the authors to describe a category of self-supervised pre-training solutions that we argue are …
Enhancing Resilience in Distributed ML Inference Pipelines for Edge Computing
As edge computing and sensing devices continue to proliferate, distributed machine
learning (ML) inference pipelines are becoming popular for enabling low-latency, real-time …
learning (ML) inference pipelines are becoming popular for enabling low-latency, real-time …
InfoMAE: Pairing-Efficient Cross-Modal Alignment with Informational Masked Autoencoders for IoT Signals
Standard multimodal self-supervised learning (SSL) algorithms regard cross-modal
synchronization as implicit supervisory labels during pretraining, thus posing high …
synchronization as implicit supervisory labels during pretraining, thus posing high …