![[2602.14481] On the Rate-Distortion-Complexity Tradeoff for Semantic Communication](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.14481] On the Rate-Distortion-Complexity Tradeoff for Semantic Communication
Summary
This paper explores the rate-distortion-complexity tradeoff in semantic communication, proposing a framework that balances semantic distance and computational complexity in deep learning models.
Why It Matters
Understanding the tradeoff between rate, distortion, and complexity is crucial for optimizing semantic communication systems. This research addresses a significant gap in current methodologies by integrating computational constraints, which can enhance the efficiency of communication technologies in resource-limited environments.
Key Takeaways
- Introduces a rate-distortion-complexity framework for semantic communication.
- Demonstrates a fundamental tradeoff among achievable rate, semantic distance, and model complexity.
- Provides theoretical results for Gaussian and binary semantic sources.
- Validates the framework through experiments on real-world datasets.
- Offers insights for designing efficient systems under computational constraints.
Computer Science > Information Theory arXiv:2602.14481 (cs) [Submitted on 16 Feb 2026] Title:On the Rate-Distortion-Complexity Tradeoff for Semantic Communication Authors:Jingxuan Chai, Yong Xiao, Guangming Shi View a PDF of the paper titled On the Rate-Distortion-Complexity Tradeoff for Semantic Communication, by Jingxuan Chai and Yong Xiao and Guangming Shi View PDF HTML (experimental) Abstract:Semantic communication is a novel communication paradigm that focuses on conveying the user's intended meaning rather than the bit-wise transmission of source signals. One of the key challenges is to effectively represent and extract the semantic meaning of any given source signals. While deep learning (DL)-based solutions have shown promising results in extracting implicit semantic information from a wide range of sources, existing work often overlooks the high computational complexity inherent in both model training and inference for the DL-based encoder and decoder. To bridge this gap, this paper proposes a rate-distortion-complexity (RDC) framework which extends the classical rate-distortion theory by incorporating the constraints on semantic distance, including both the traditional bit-wise distortion metric and statistical difference-based divergence metric, and complexity measure, adopted from the theory of minimum description length and information bottleneck. We derive the closed-form theoretical results of the minimum achievable rate under given constraints on semantic d...
