mlsanityData to 14 Oct 2026

Bit Error and Block Error Rate Training for ML-Assisted Communication

arXiv 2210.14103 ↗First posted 25 Oct 2022cs.IT cs.LG eess.SPcited by 6

Not on the map. Why?

Abstract (arXiv)

Even though machine learning (ML) techniques are being widely used in communications, the question of how to train communication systems has received surprisingly little attention. In this paper, we show that the commonly used binary cross-entropy (BCE) loss is a sensible choice in uncoded systems, e.g., for training ML-assisted data detectors, but may not be optimal in coded systems. We propose new loss functions targeted at minimizing the block error rate and SNR deweighting, a novel method that trains communication systems for optimal performance over a range of signal-to-noise ratios. The utility of the proposed loss functions as well as of SNR deweighting is shown through simulations in NVIDIA Sionna.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 16 of its 32 references that are in the map. First posted on arXiv 25 Oct 2022.

Builds on

16 of 32 references.

References · 16

  1. 2022Sionna: An Open-Source Library for Next-Generation Physical Layer Researchcited by 145
  2. 2018Overfitting or perfect fitting? Risk bounds for classification and regression rules that interpolatecited by 99
  3. 2017Deep Learning-Based Communication Over the Aircited by 72
  4. 2017Deep Learning Methods for Improved Decoding of Linear Codescited by 66
  5. 2018Learning to Detectcited by 63
  6. 2020DeepRx: Fully Convolutional Deep Learning Receivercited by 52
  7. 2019Deep Unfolding for Communications Systems: A Survey and Some New Directionscited by 49
  8. 2015Explaining the Success of AdaBoost and Random Forests as Interpolating Classifierscited by 45
  9. 2018Model-free Training of End-to-end Communication Systemscited by 30
  10. 2019Adaptive Neural Signal Detection for Massive MIMOcited by 27
  11. 2019Trainable Communication Systems: Concepts and Prototypecited by 22
  12. 20185G New Radio: Unveiling the Essentials of the Next Generation Wireless Access Technologycited by 16
4 more
  1. 2019Joint Learning of Geometric and Probabilistic Constellation Shapingcited by 13
  2. 2022Benchmarking and Interpreting End-to-end Learning of MIMO and Multi-User Communicationcited by 7
  3. 2019What Can Machine Learning Teach Us about Communications?cited by 4
  4. 2021Soft-Output Joint Channel Estimation and Data Detection using Deep Unfoldingcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 1; 2023: 1; 2024: 1; 2025: 2; 2026: 1.

6 papers.

  1. 2022DUIDD: Deep-Unfolded Interleaved Detection and Decoding for MIMO Wireless Systemscited by 1
  2. 2023SICNN: Soft Interference Cancellation Inspired Neural Network Equalizerscited by 1
  3. 2024Modular Neural Wiretap Codes for Fading Channelscited by 0
  4. 2025Learning Linear Block Codes with Gradient Quantizationcited by 0
  5. 2025Forward-Forward Autoencoder Architectures for Energy-Efficient Wireless Communicationscited by 0
  6. 2026Learning-to-Transition for Large-scale and High-Order MIMO Detectioncited by 0