Policy-Guided MCTS for near Maximum-Likelihood Decoding of Short Codes

2025-11-12

AI Summary

To address the high computational overhead and latency induced by Gaussian elimination in near-maximum-likelihood decoding (Near-MLD) of short codes, this paper proposes a policy-guided Monte Carlo Tree Search (MCTS) decoder. The method searches for test error patterns in the information bit space, generates candidate codewords via re-encoding, and employs a neural network policy model—trained end-to-end by MCTS—to dynamically guide the search trajectory, augmented with an early-stopping mechanism to accelerate convergence. This work is the first to apply policy-guided MCTS to short-code Near-MLD, eliminating the need for Gaussian elimination while maintaining decoding efficiency. Experimental results demonstrate that, compared to ordered statistics decoding (OSD), the proposed decoder reduces search complexity by 95% and significantly lowers decoding latency at high signal-to-noise ratios, achieving bit-error-rate performance within close proximity to maximum-likelihood decoding.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations

Abstract

In this paper, we propose a policy-guided Monte Carlo Tree Search (MCTS) decoder that achieves near maximum-likelihood decoding (MLD) performance for short block codes. The MCTS decoder searches for test error patterns (TEPs) in the received information bits and obtains codeword candidates through re-encoding. The TEP search is executed on a tree structure, guided by a neural network policy trained via MCTS-based learning. The trained policy guides the decoder to find the correct TEPs with minimal steps from the root node (all-zero TEP). The decoder outputs the codeword with maximum likelihood when the early stopping criterion is satisfied. The proposed method requires no Gaussian elimination (GE) compared to ordered statistics decoding (OSD) and can reduce search complexity by 95% compared to non-GE OSD. It achieves lower decoding latency than both OSD and non-GE OSD at high SNRs.

Problem

• Achieving near maximum-likelihood decoding for short block codes
• Reducing search complexity by 95% compared to non-GE OSD
• Eliminating Gaussian elimination requirements in decoding process

Innovation

• Policy-guided MCTS decoder for short block codes
• Neural network policy trained via MCTS-based learning
• No Gaussian elimination required, reduces search complexity