`

Timezone: »

 
Poster
Hyper-Graph-Network Decoders for Block Codes
Eliya Nachmani · Lior Wolf

Wed Dec 11 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #175
in Probabilistic Methods -- Belief Propagation »

Neural decoders were shown to outperform classical message passing techniques for short BCH codes. In this work, we extend these results to much larger families of algebraic block codes, by performing message passing with graph neural networks. The parameters of the sub-network at each variable-node in the Tanner graph are obtained from a hypernetwork that receives the absolute values of the current message as input. To add stability, we employ a simplified version of the arctanh activation that is based on a high order Taylor approximation of this activation function. Our results show that for a large number of algebraic block codes, from diverse families of codes (BCH, LDPC, Polar), the decoding obtained with our method outperforms the vanilla belief propagation method as well as other learning techniques from the literature.

Keywords: Supervised Deep Networks · Meta-Learning · Algorithms · Applications -> Denoising; Applications -> Signal Processing; Deep Learning
[ Paper]  [ Slides] 

Author Information

Eliya Nachmani (Tel Aviv University and Facebook AI Research)
Lior Wolf (Facebook AI Research)

More from the Same Authors