Semantic Meta-Split Learning: A TinyML Scheme for Few-Shot Wireless Image Classification
This work addresses privacy and efficiency issues in semantic communication for few-shot image classification, though it is incremental as it integrates existing techniques.
The paper tackles the challenges of computational complexity, data availability, and privacy in semantic communication for wireless image classification by proposing a TinyML framework that combines split-learning and meta-learning. It achieves a 20% gain in classification accuracy with fewer data points and lower training energy consumption.
Semantic and goal-oriented (SGO) communication is an emerging technology that only transmits significant information for a given task. Semantic communication encounters many challenges, such as computational complexity at end users, availability of data, and privacy-preserving. This work presents a TinyML-based semantic communication framework for few-shot wireless image classification that integrates split-learning and meta-learning. We exploit split-learning to limit the computations performed by the end-users while ensuring privacy-preserving. In addition, meta-learning overcomes data availability concerns and speeds up training by utilizing similarly trained tasks. The proposed algorithm is tested using a data set of images of hand-written letters. In addition, we present an uncertainty analysis of the predictions using conformal prediction (CP) techniques. Simulation results show that the proposed Semantic-MSL outperforms conventional schemes by achieving 20 % gain on classification accuracy using fewer data points, yet less training energy consumption.