Research map / Graph learning / Graph algorithms and clustering
Learning to Decode: Reinforcement Learning for Decoding of Sparse Graph-Based Channel Codes
Salman Habib, Allison Beemer, Joerg Kliewer
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use learned sequential check-node scheduling instead of flooding BP | LDPC decoding · (3,6)-regular and (3,7) AB-LDPC | BER | Improved |
| Use learned sequential check-node scheduling instead of flooding BP | LDPC decoding · (3,6)-regular LDPC, SNR 0.5 dB | CN-to-VN messages propagated: 27040 -> 264 | Improved |
| Use cycle-maximized clusters rather than contiguous clusters (ablation) | Clustered Q-learning (MQC) · Both LDPC codes | BER | Improved |
| Use cycle-maximized clusters rather than randomly ordered clusters (ablation) | Clustered Q-learning (MQR) · Both LDPC codes | BER | Improved |
| Use cycle-maximized clusters rather than contiguous clusters (ablation) | Clustered Q-learning (MQC) · (3,6)-regular LDPC, SNR 0.5 dB | CN-to-VN messages propagated: 351 -> 264 | Improved |
| Use cycle-maximized clusters rather than randomly ordered clusters (ablation) | Clustered Q-learning (MQR) · (3,6)-regular LDPC, SNR 0.5 dB | CN-to-VN messages propagated: 310 -> 264 | Improved |
| Use cycle-maximized clusters rather than contiguous clusters (ablation) | Clustered Q-learning (MQC) · (3,7) AB-LDPC, SNR 0.5 dB | CN-to-VN messages propagated: 411 -> 367 | Improved |
| Use cycle-maximized clusters rather than randomly ordered clusters (ablation) | Clustered Q-learning (MQR) · (3,7) AB-LDPC, SNR 0.5 dB | CN-to-VN messages propagated: 368 -> 367 | No clear effect |
| Use Q-learning rather than Thompson Sampling for sequential scheduling | Sequential BP decoding · LDPC codes | BER | Improved |
| Use sequential RL-based scheduling rather than residual-based NS | LDPC decoding · Short LDPC codes | message-passing complexity | Improved |
Closest papers by meaning
- Friendly Attacks to Improve Channel Coding Reliability (AAAI 2024 · similarity 0.83)
- Learning Linear Block Error Correction Codes (ICML 2024 · similarity 0.81)
- Tokenized Bandit for LLM Decoding and Alignment (ICML 2025 · similarity 0.81)
- Learning To Draft: Adaptive Speculative Decoding with Reinforcement Learning (ICLR 2026 · similarity 0.81)
- Controlled Decoding from Language Models (ICML 2024 · similarity 0.81)
- A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability (NeurIPS 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
