Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 134 tok/s
Gemini 2.5 Pro 41 tok/s Pro
GPT-5 Medium 15 tok/s Pro
GPT-5 High 26 tok/s Pro
GPT-4o 82 tok/s Pro
Kimi K2 198 tok/s Pro
GPT OSS 120B 436 tok/s Pro
Claude Sonnet 4.5 37 tok/s Pro
2000 character limit reached

DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning (2402.08864v2)

Published 14 Feb 2024 in cs.IT, cs.LG, and math.IT

Abstract: Progress in designing channel codes has been driven by human ingenuity and, fittingly, has been sporadic. Polar codes, developed on the foundation of Arikan's polarization kernel, represent the latest breakthrough in coding theory and have emerged as the state-of-the-art error-correction code for short-to-medium block length regimes. In an effort to automate the invention of good channel codes, especially in this regime, we explore a novel, non-linear generalization of Polar codes, which we call DeepPolar codes. DeepPolar codes extend the conventional Polar coding framework by utilizing a larger kernel size and parameterizing these kernels and matched decoders through neural networks. Our results demonstrate that these data-driven codes effectively leverage the benefits of a larger kernel size, resulting in enhanced reliability when compared to both existing neural codes and conventional Polar codes.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (52)
  1. 3GPP. Multiplexing and channel coding. Technical Specification 38.212, 3rd Generation Partnership Project (3GPP), 2018. Release 15.
  2. Large kernel polar codes with efficient window decoding. IEEE Transactions On Vehicular Technology, 69(11):14031–14036, 2020.
  3. Data-driven polar codes for unknown channels with and without memory. In 2023 IEEE International Symposium on Information Theory (ISIT), pp.  1890–1895. IEEE, 2023.
  4. Nested construction of polar codes via transformers. arXiv preprint arXiv:2401.17188, 2024.
  5. Arikan, E. Channel polarization: A method for constructing capacity-achieving codes for symmetric binary-input memoryless channels. IEEE Transactions on information Theory, 55(7):3051–3073, 2009.
  6. Arıkan, E. From sequential decoding to channel polarization and back again. arXiv preprint arXiv:1908.09594, 2019.
  7. Near shannon limit error-correcting coding and decoding: Turbo-codes. 1. In Proceedings of ICC’93-IEEE International Conference on Communications, volume 2, pp.  1064–1070. IEEE, 1993.
  8. Design of polar codes in 5g new radio. IEEE Communications Surveys & Tutorials, 23(1):29–40, 2020.
  9. Scaling deep learning-based decoding of polar codes via partitioning. In GLOBECOM 2017-2017 IEEE global communications conference, pp.  1–6. IEEE, 2017a.
  10. Combining belief propagation and successive cancellation list decoding of polar codes on a gpu platform. In 2017 IEEE international conference on acoustics, speech and signal processing (ICASSP), pp.  3664–3668. IEEE, 2017b.
  11. Deepic: Coding for interference channels via deep learning. arXiv preprint arXiv:2108.06028, 2021.
  12. Turbo autoencoder with a trainable interleaver. In ICC 2022-IEEE International Conference on Communications, pp.  3886–3891. IEEE, 2022a.
  13. Inventing codes for channels with active feedback via deep learning. IEEE Journal on Selected Areas in Information Theory, 3(3):574–586, 2022b.
  14. Cyclically equivariant neural decoders for cyclic codes. arXiv preprint arXiv:2105.05540, 2021.
  15. Denoising diffusion error correction codes. arXiv preprint arXiv:2209.13533, 2022a.
  16. Error correction code transformer. arXiv preprint arXiv:2203.14966, 2022b.
  17. Interpreting deep-learned error-correcting codes. In 2022 IEEE International Symposium on Information Theory (ISIT), pp.  2457–2462, 2022. doi: 10.1109/ISIT50566.2022.9834599.
  18. Neural successive cancellation decoding of polar codes. In 2018 IEEE 19th international workshop on signal processing advances in wireless communications (SPAWC), pp.  1–5. IEEE, 2018.
  19. Deep learning based communication over the air. IEEE Journal of Selected Topics in Signal Processing, 12(1):132–143, 2017.
  20. Deep learning-based polar code design. In 2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton), pp.  177–183. IEEE, 2019.
  21. On the scaling exponent of binary polarization kernels. In 2014 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton), pp.  797–804. IEEE, 2014.
  22. Binary linear codes with optimal scaling: Polar codes with large kernels. IEEE Transactions on Information Theory, 67(9):5693–5710, 2020.
  23. Polar list decoding for large polarization kernels. In 2021 IEEE Globecom Workshops (GC Wkshps), pp.  1–6. IEEE, 2021.
  24. Tinyturbo: Efficient turbo decoders on edge. In 2022 IEEE International Symposium on Information Theory (ISIT), pp.  2797–2802. IEEE, 2022.
  25. Crisp: Curriculum based sequential neural decoders for polar code family. In International Conference on Machine Learning, pp.  12823–12845. PMLR, 2023.
  26. Binarized neural networks. Advances in neural information processing systems, 29, 2016.
  27. Reed-muller subcodes: Machine learning-aided design of efficient soft recursive decoding. In 2021 IEEE International Symposium on Information Theory (ISIT), pp.  1088–1093. IEEE, 2021a.
  28. Productae: Towards training larger channel codes based on neural product codes. arXiv preprint arXiv:2110.04466, 2021b.
  29. Deepturbo: Deep turbo decoder. In 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp.  1–5. IEEE, 2019a.
  30. Turbo autoencoder: Deep learning based channel codes for point-to-point communication channels. In Advances in Neural Information Processing Systems, pp.  2758–2768, 2019b.
  31. Joint channel coding and modulation via deep learning. In 2020 IEEE 21st International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp.  1–5. IEEE, 2020a.
  32. Feedback turbo autoencoder. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.  8559–8563. IEEE, 2020b.
  33. Deepcode: Feedback codes via deep learning. Advances in neural information processing systems, 31, 2018a.
  34. Communication algorithms via deep learning. arXiv preprint arXiv:1805.09317, 2018b.
  35. Robust non-linear feedback coding via power-constrained deep learning. arXiv preprint arXiv:2304.13178, 2023.
  36. Polar codes: Characterization of exponent, bounds, and constructions. IEEE Transactions on Information Theory, 56(12):6253–6264, 2010.
  37. Learning to construct nested polar codes: An attention-based set-to-element model. IEEE Communications Letters, 25(12):3898–3902, 2021.
  38. Construction of polar codes with reinforcement learning. IEEE Transactions on Communications, 70(1):185–198, 2021.
  39. Ko codes: inventing nonlinear encoding and decoding for reliable wireless communication via deep-learning. In International Conference on Machine Learning, pp.  7368–7378. PMLR, 2021.
  40. Design of compactly specified polar codes with dynamic frozen bits based on reinforcement learning. IEEE Transactions on Communications, 2023.
  41. Hyper-graph-network decoders for block codes. Advances in Neural Information Processing Systems, 32:2329–2339, 2019.
  42. Learning to decode linear codes using deep learning. In 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton), pp.  341–346. IEEE, 2016.
  43. Crc-aided decoding of polar codes. IEEE communications letters, 16(10):1668–1671, 2012.
  44. All you need is feedback: Communication with block attention feedback codes. IEEE Journal on Selected Areas in Information Theory, 3(3):587–602, 2022.
  45. Do not interfere but cooperate: A fully learnable code design for multi-access channels with feedback. In 2023 12th International Symposium on Topics in Coding (ISTC), pp.  1–5. IEEE, 2023a.
  46. Feedback is good, active feedback is better: Block attention active feedback codes. In ICC 2023-IEEE International Conference on Communications, pp.  6652–6657. IEEE, 2023b.
  47. List autoencoder: Towards deep learning based reliable transmission over noisy channels. In GLOBECOM 2022-2022 IEEE Global Communications Conference, pp.  1454–1459. IEEE, 2022.
  48. Deep extended feedback codes. arXiv preprint arXiv:2105.01365, 2021.
  49. Shannon, C. E. A mathematical theory of communication. The Bell system technical journal, 27(3):379–423, 1948.
  50. Window processing of binary polarization kernels. IEEE Transactions on Communications, 69(7):4294–4305, 2021.
  51. Learning to decode ldpc codes with finite-alphabet message passing. In 2018 Information Theory and Applications Workshop (ITA), pp.  1–9. IEEE, 2018.
  52. Rate-matched turbo autoencoder: A deep learning based multi-rate channel autoencoder. In ICC 2023-IEEE International Conference on Communications, pp.  6355–6360. IEEE, 2023.
Citations (3)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up