SC-GIR: Goal-oriented Semantic Communication via Invariant Representation Learning
This work addresses efficiency and task-agnostic utility in machine-to-machine communication systems, representing an incremental advancement in semantic communication methods.
The paper tackles the problem of redundant data exchange and reliance on labeled datasets in goal-oriented semantic communication for image transmission by proposing SC-GIR, a framework that uses self-supervised learning to extract invariant representations, resulting in nearly 10% performance improvement over baselines and over 85% classification accuracy for compressed data under various SNR conditions.
Goal-oriented semantic communication (SC) aims to revolutionize communication systems by transmitting only task-essential information. However, current approaches face challenges such as joint training at transceivers, leading to redundant data exchange and reliance on labeled datasets, which limits their task-agnostic utility. To address these challenges, we propose a novel framework called Goal-oriented Invariant Representation-based SC (SC-GIR) for image transmission. Our framework leverages self-supervised learning to extract an invariant representation that encapsulates crucial information from the source data, independent of the specific downstream task. This compressed representation facilitates efficient communication while retaining key features for successful downstream task execution. Focusing on machine-to-machine tasks, we utilize covariance-based contrastive learning techniques to obtain a latent representation that is both meaningful and semantically dense. To evaluate the effectiveness of the proposed scheme on downstream tasks, we apply it to various image datasets for lossy compression. The compressed representations are then used in a goal-oriented AI task. Extensive experiments on several datasets demonstrate that SC-GIR outperforms baseline schemes by nearly 10%,, and achieves over 85% classification accuracy for compressed data under different SNR conditions. These results underscore the effectiveness of the proposed framework in learning compact and informative latent representations.