Leveraging Overfitting for Low-Complexity and Modality-Agnostic Joint Source-Channel Coding

2025-12-24

AI Summary

Traditional joint source-channel coding (JSCC) suffers from heavy reliance on large-scale training data, modality-specific designs, high decoding complexity, and poor generalization. To address these limitations, this paper proposes Implicit-JSCC—a paradigm that operates per single-image instance via overfitting-driven implicit neural representation, directly optimizing channel symbols and an ultra-lightweight decoder (only 607 parameters, 641 multiplications per pixel), without any training dataset or pre-trained model. The method enables storage-free, cross-modal, single-shot offline encoding and repeated online decoding, inherently supporting source generalization. In high-SNR image transmission tasks, it achieves state-of-the-art (SOTA) performance while reducing decoding complexity by approximately three orders of magnitude—making it highly suitable for low-latency applications such as streaming media.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

This paper introduces Implicit-JSCC, a novel overfitted joint source-channel coding paradigm that directly optimizes channel symbols and a lightweight neural decoder for each source. This instance-specific strategy eliminates the need for training datasets or pre-trained models, enabling a storage-free, modality-agnostic solution. As a low-complexity alternative, Implicit-JSCC achieves efficient image transmission with around 1000x lower decoding complexity, using as few as 607 model parameters and 641 multiplications per pixel. This overfitted design inherently addresses source generalizability and achieves state-of-the-art results in the high SNR regimes, underscoring its promise for future communication systems, especially streaming scenarios where one-time offline encoding supports multiple online decoding.

Problem

• Develops a low-complexity, modality-agnostic joint source-channel coding method
• Eliminates need for training datasets or pre-trained models for each source
• Achieves efficient image transmission with minimal parameters and multiplications

Innovation

• Overfitted JSCC optimizes symbols and decoder per source
• Eliminates training data, storage-free, modality-agnostic solution
• Low-complexity with 1000x fewer decodes, 607 parameters used