Explainable AI for inverse-design of optical network on chip routers
-
Jotiram K. Deshmukh
,
Milind P. Gajare
Abstract
Inverse design uses advanced AI capabilities to autonomously create the best geometries and configurations for an optical network-on-chip router. Here, we use reinforcement learning to facilitate the design process, where the system learns over time what architectures best suit the target criteria (latency, signal loss, area, throughput) to achieve the best functioning router. However, while reinforcement learning is an effective means of achieving desired output, the reinforcement learning created is often complex and non-interpretable for human engineers. Therefore, we use explainable artificial intelligence methods to make the final interpretations more interpretable and justifiable. Explainable artificial intelligence helps to explain why each move was made during the design process. Thus, reinforcement learning-inverse-designed optical network-on-chip routers will perform better under desired metrics with explainable artificial intelligence providing human designers a level of explainability and justification for human trust and verification for artificial intelligence generated things.
-
Research ethics: Not applicable.
-
Informed consent: Not applicable.
-
Author contributions: The authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Use of Large Language Models, AI and Machine Learning Tools: None declared.
-
Conflict of interest: The authors state no conflict of interest.
-
Research funding: None declared.
-
Data availability: Not applicable.
References
1. Li, J. The intelligent design of silicon photonic devices. Nanophotonics 2024;12:2301337.Search in Google Scholar
2. Soref, R. Review of ONoC architectures. IEEE J Quant Electron 2024;60:1–12.Search in Google Scholar
3. Kojima, K, Tang, Y, Koike, T, Wang, Y, Jha, D, Parsons, K, et al.. Inverse design of nanophotonic devices with deep learning. Asia Commun Photonics 2020. https://doi.org/10.1364/ACPC.2020.Su1A.1.Search in Google Scholar
4. Molesky, S, Lin, Z, Piggott, A, Jin, W, Vučković, J, Rodriguez, A. Inverse design in nano-photonics. Nat Photonics 2020;12:659–70.10.1038/s41566-018-0246-9Search in Google Scholar
5. Moon, S, Kang, J, Yeo, J-S. Review of generative models for the inverse design of nanophotonic metasurfaces. Appl Sci Converg Technol 2023;32:141–50.10.5757/ASCT.2023.32.6.141Search in Google Scholar
6. Deng, Y, Ren, S, Malof, J, Padilla, W. Deep inverse photonic design: a tutorial. Opt Express 2022;30:2538–65.10.2139/ssrn.4204784Search in Google Scholar
7. Jensen, J, Sigmund, O. Topology optimization for nano-photonic devices. Laser Photon Rev 2011;5:308–21.10.1002/lpor.201000014Search in Google Scholar
8. Hughes, T, Minkov, M, Williamson, A, Fan, S. Adjoint method and inverse design for nonlinear nanophotonic devices. ACS Photonics 2018;5:4781–7.10.1021/acsphotonics.8b01522Search in Google Scholar
9. Zhang, X, Xie, W, Leong, C, Gao, F. Compact inverse-designed tilted waveguide crossing. Opt Lett 2025;50:1889–92.10.1364/OL.553929Search in Google Scholar PubMed
10. Peurifoy, J, Shen, Y, Jing, L, Picard, M, Doran, A, Fan, S. Nanophotonic particle simulation and inverse design using artificial neural networks. Sci Adv 2018;4:eaar4206.10.1126/sciadv.aar4206Search in Google Scholar PubMed PubMed Central
11. Ma, H. Variational autoencoders for inverse design of photonics. Opt Express 2020;28:25453–65.10.1364/OE.389474Search in Google Scholar PubMed
12. Jiang, J, Sell, D, Hoyer, S, Hickey, J, Yang, J, Fan, J. Generative adversarial networks for photonic structure design. Nanophotonics 2021;10:4885–98.Search in Google Scholar
13. Verma, R, Saxena, A, Pandey, V, Gupta, A. Diffusion models for inverse photonic design. Opt Express 2024;32:4012–26.Search in Google Scholar
14. Hughes, T, Aspuru-Guzik, A. PHORCED: reinforcement learning for photonic design. Nanophotonics 2021;10:1259–70.Search in Google Scholar
15. Witt, D, Young, J, Chrostowski, L. Reinforcement learning approaches for photonics. APL Photonics 2023;8:106101.10.1063/5.0159928Search in Google Scholar
16. Yeung, C, Pham, B, Tsai, R, Fountaine, KT, Raman, AP. DeepAdjoint: an all-in-one photonic inverse design framework integrating data-driven machine learning with optimization algorithms. arXiv preprint arXiv:2209.15434 2022.10.1021/acsphotonics.2c00968Search in Google Scholar
17. Seo, D, Um, S, Lee, S, Ye, JC, Chung, H. Physics-guided and fabrication-aware inverse design of photonic devices using diffusion models. arXiv preprint arXiv:2504.17077 2025.Search in Google Scholar
18. Seo, D, Um, S, Lee, S, Chul Ye, J, Chung, H. Physics-guided and fabrication-aware inverse design of photonic devices using diffusion models. ACS Photonics 2023;10:1234–44.Search in Google Scholar
19. Yuan, M, Yang, G, Song, S, Zhou, L, Minasian, R, Yi, X. Inverse design of a nano-photonic wavelength demultiplexer with a deep neural network approach. Opt Express 2022;29:12345–56.10.1364/OE.462038Search in Google Scholar PubMed
20. Zhu, Y. Broadband WDM router design using inverse methods. IEEE Photon Technol Lett 2023;35:502–5.10.1109/LPT.2023.3298841Search in Google Scholar
21. He, X, Sun, D, Chen, J, Shi, Y. Inverse designed grating coupler with low loss and high bandwidth on LNOI platform. IEEE Photonics J 2021;16:1–5.10.1109/JPHOT.2024.3351199Search in Google Scholar
22. Zhang, L. Compact waveguide crossings using topology optimization. J Lightwave Technol 2021;39:2345–52.Search in Google Scholar
23. Schubert, F, Mahlau, Y, Bethmann, K, Hartmann, F, Caspary, R, Ostermann, J, et al.. Quantized inverse design for photonic integrated circuits. arXiv:2407.10273v2 2024;1–16. https://doi.org/10.48550/arXiv.2407.10273.Search in Google Scholar
24. Schuber, M, Cheung, A, Williamson, I, Spyra, A, Alexander, D. Inverse design of photonics under strict fabrication constraints. ACS Photonics 2022;9:2327–36.10.1021/acsphotonics.2c00313Search in Google Scholar
25. Lee, AR. Explainable ML for photonic inverse design. ACS Photonics 2022;9:2234–41.Search in Google Scholar
26. Gupta, K. XAI approaches in nanophotonics. IEEE Access 2023;11:87654–62.Search in Google Scholar
27. Park, J. Diffusion-driven inverse design of broadband photonic devices. In: CLEO conference proceedings. Charlotte, North Carolina, USA: CLCO; 2024.Search in Google Scholar
28. Lutheran, A, Das, S. Latent diffusion for topology-constrained photonics. Optica 2024;11:145–53.Search in Google Scholar
29. Goel, S, Leedumrongwatthanakun, S, Valencia, NH, McCutcheon, W, Tavakoli, A, Conti, C, et al.. Inverse design of high-dimensional quantum optical circuits in a complex medium. Nat Phys 2024;20:232–9.10.1038/s41567-023-02319-6Search in Google Scholar
30. Lenzini, F. Integrated quantum routers by inverse design. npj Quant Inf 2022;8.Search in Google Scholar
31. Raza, S, Hammood, M, Jaeger, N, Chrostowski, L. Fabrication-aware inverse design with shape optimization for photonic integrated circuits. Opt Lett 2024;50:117–20.10.1364/OL.543961Search in Google Scholar PubMed
32. Johnson, A. Fabrication-in-the-loop photonic inverse design. Opt Express 2023;31:1150–62.Search in Google Scholar
33. Kang, C, Park, C, Lee, M, Kang, J. Large-scale photonic inverse design: computational challenges and breakthroughs. Nanophotonics 2024;13:3766–92.10.1515/nanoph-2024-0127Search in Google Scholar PubMed PubMed Central
34. Karahan, EA, Ulusoy, AC, Mousavi, SH, Joshi, S, Keskin, M, Bagci, H. Deep-learning enabled generalized inverse design of multi-port radio-frequency and sub-terahertz passives and integrated circuits. Nat Commun 2024;15:10734. https://doi.org/10.1038/s41467-024-54178-1.Search in Google Scholar PubMed PubMed Central
35. Pan, Z-Y, Pan, X-M. Deep learning and adjoint method accelerated inverse design in photonics: a review. Photonics 2023;10:852. https://doi.org/10.3390/photonics10080852.Search in Google Scholar
36. Adadi, A, Berrada, M. Peeking inside the black-box: a survey on explainable artificial intelligence (XAI). IEEE Access 2018;6:52138–60.10.1109/ACCESS.2018.2870052Search in Google Scholar
37. Fan, L, IEEE Standards Association. IEEE standard for explainable artificial intelligence—IEEE 2894-2024. IEEE 2024. https://standards.ieee.org/ieee/2894/11296.Search in Google Scholar
38. Tjoa, E, Guan, C. A survey on explainable artificial intelligence (XAI): toward medical XAI. IEEE Transact Neural Networks Learn Syst 2024;32:4793–813.10.1109/TNNLS.2020.3027314Search in Google Scholar PubMed
39. Zhang, R, Hamad́ı, HA, Damiani, E, Yeun, CY, Taher, F. Explainable artificial intelligence applications in cyber security: state-of-the-art in research. IEEE Access 2022;10:93104–39. https://doi.org/10.1109/ACCESS.2022.3204051.Search in Google Scholar
40. Martins, K, de Almeida, T, Cardoso, E, Nunes, L. Explainable artificial intelligence (XAI): a systematic literature review on taxonomies and applications in finance. IEEE Access 2024;12:618–29.10.1109/ACCESS.2023.3347028Search in Google Scholar
41. Ribeiro, M, Singh, S, Guestrin, C. ‘Why should I trust you?’: explaining the predictions of any classifier. In: Proceedings of the ACM SIGKDD international conference on knowledge discovery and data mining (KDD), San Francisco California. ACM; 2016.10.1145/2939672.2939778Search in Google Scholar
42. Alizadehsani, R, Oyelere, SS, Hussain, S, Jagatheesaperumal, SK, Calixto, RR, Rahouti, M, et al.. Explainable artificial intelligence for drug discovery and development — a comprehensive survey. IEEE Access 2024;12:35796–812. https://doi.org/10.1109/ACCESS.2024.3373195.Search in Google Scholar
43. Kalasampath, K, KN, S, Sajeev, S. A literature review on applications of explainable artificial intelligence (XAI). IEEE Access 2025;13. https://doi.org/10.1109/ACCESS.2025.3546681.Search in Google Scholar
44. Baniecki, H, Biecek, P. Adversarial attacks and defenses in explainable artificial intelligence: a survey. arXiv Preprint 2023.10.1016/j.inffus.2024.102303Search in Google Scholar
45. Ahmed, S, Kaiser, MS, Shahadat Hossain, M, Andersson, K. Comparative analysis of LIME and SHAP interpreters with explainable ML-based diabetes predictions. IEEE Access 2024;13:37370–88. https://doi.org/10.1109/ACCESS.2024.3422319.Search in Google Scholar
46. Yeung, C, Ho, D, Pham, B, Fountaine, KT, Zhang, Z, Levy, K, et al.. Enhancing adjoint optimization-based photonic inverse design with explainable machine learning. ACS Photonics 2022;9:1577–85. https://doi.org/10.1021/acsphotonics.1c01636.Search in Google Scholar
47. Pira, L, Antony, A, Prathap, N, Peace, D, Romero, J. Enhanced photonic chip design via interpretable machine learning techniques. arXiv preprint arXiv:2505.09266 2025. [Online]. Available: https://arxiv.org/abs/2505.09266.Search in Google Scholar
© 2025 Walter de Gruyter GmbH, Berlin/Boston
