Relaxed protograph entries as Bernoulli edge probabilities enable deterministic DE-based gradient descent that yields LDPC base graphs outperforming 5G references of the same size.
Learning linear block codes with gradient quantization,
2 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
2
Pith papers citing it
1
external citations · external index
years
2026 2representative citing papers
Product networks with stochastic sparsity (Bernoulli p_e ≤ 1/N) enable gradient descent to learn high-dimensional parity functions with theoretical convergence guarantees and polynomial scaling up to N=100,000.
citing papers explorer
-
Learning LDPC codes with quantized density evolution over relaxed protographs
Relaxed protograph entries as Bernoulli edge probabilities enable deterministic DE-based gradient descent that yields LDPC base graphs outperforming 5G references of the same size.
-
Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent
Product networks with stochastic sparsity (Bernoulli p_e ≤ 1/N) enable gradient descent to learn high-dimensional parity functions with theoretical convergence guarantees and polynomial scaling up to N=100,000.