edfas.org 27 ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 25 NO. 2 18. C.E. Shannon: https://doi.org/10.1109/T-AIEE.1938.5057767. 19. J. de Kleer and B.C. Williams: https://doi.org/10.1016/0004-3702 (87)90063-4. 20. R. Reiter: “A Theory of Diagnosis from First Principles,” Artif. Intell., 32.1, p. 57–95, 1987. 21. C. Burmer, et al.: https://doi.org/10.31399/asm.cp.istfa2005p0395. 22. S. Holst and H.-J. Wunderlich: https://doi.org/10.1007/s10836- 009-5109-3. 23. C. Hora and S. Eichenberger: https://doi.org/10.31399/asm. cp.istfa2004p0047. 24. C.L. Henderson and J.M. Soden: https://doi.org/10.1109/ RELPHY.1991.146003. 25. T. Tsujide, et al.: https://doi.org/10.1109/RELPHY.1993.283326. 26. Hongjian Wang et al.: https://doi.org/10.1109/IPFA47161. 2019.8984840. 27. A. Safont-Andreu, C. Burmer, and K. Schekotihin: https://doi.org/ 10.31399/asm.cp.istfa2021p0023. 28. T. Berners-Lee, J. Hendler, and O. Lassila: “The Semantic Web,” Scientific American, 2001, 284.5, p. 34–43. 29. M. Kučko: “LotTraveler: Workflow Management in Failure Analysis,” PhD thesis, Alpen-Adria-University Klagenfurt, 2022, https://digital. obvsg.at/urn/urn:nbn:at:at-ubk:1-45251. 30. I.J. Goodfellow, Y. Bengio, and A.C. Courville: Deep Learning, MIT Press, 2016. 31. J. MacQueen: “Some Methods for Classification and Analysis of Multivariate Observations,” Berkeley Symp. Math. Statist. Prob., p. 281–297, 1967. 32. P. Tassel, M. Gebser, and K. Schekotihin: “An End-to-End Reinforcement Learning Approach for Job-Shop Scheduling Problems Based on Constraint Programming,” ICAPS, AAAI Press, to be published 2023. 33. G. James, et al.: An Introduction to Statistical Learning, 112, Springer, 2021. 34. K. Sparck Jones: “A Statistical Interpretation of Term Specificity and its Application in Retrieval,” J. of Documentation, 1972, 28(1), p. 11–21. 35. S. Hochreiter and J. Schmidhuber: https://doi.org/10.1162/neco. 1997.9.8.1735. 36. A. Vaswani, et al.: “Attention is All you Need,” NIPS 2017, p. 5998–6008, 2017. 37. S. Kamm, et al.: https://doi.org/10.1109/IPFA53173.2021.9617401. 38. L. Zhao, et al.: https://doi.org/10.1109/IPFA.2018.8452176. 39. X. Hong, et al.: https://doi.org/10.1109/IPFA53173.2021.9617311. 40. M.M. Bronstein, et al.: https://doi.org/10.48550/arXiv.2104.13478. 41. S. Lobnig, C. Burmer, and K. Schekotihin: https://doi.org/10.1109/ IPFA55383.2022.9915737. 42. M.A.M. Sathiaseelan, et al.: https://doi.org/10.31399/asm.cp. istfa2021p0012. 43. M. Kögel, S. Brand, and F. Altmann: https://doi.org/10.31399/asm. cp.istfa2019p0035. 44. S. Brand, et al.: https://doi.org/10.1016/j.microrel.2010.07.139. 45. C. Grabner, et al.: https://doi.org/10.31399/asm.cp.istfa2022p0028. 46. F. Platter, et al.: https://doi.org/10.31399/asm.cp.istfa2021p0001. 47. I. Beltagy, K. Lo, and A. Cohan: http://dx.doi.org/10.18653/v1/ D19-1371. 48. K. Ezukwoke, et al.: https://doi.org/10.1109/CASE49439.2021.9551492. 49. H. Lu, et al.: https://doi.org/10.31399/asm.cp.istfa2020p0020. 50. C. Bergès, et al.: https://doi.org/10.1109/IPFA55383.2022.9915769. 51. T.B. Brown, et al.: “Language Models are Few-Shot Learners,” NeurIPS, 2020. 52. X. Liu, et al.: https://doi.org/10.1109/ACCESS.2020.3045078. APPENDIX Methodology. To collect all the relevant works, the search methodology was divided into two main phases. First, relevant conferences in FA were reviewed for AI-related papers. Next, the search was extended using Google Scholar and the IEEE Xplore, which resulted in 51 selected papers describing AI applications in FA. Regarding papers from specific FA conferences, a search of International Symposium for Testing and Failure Analysis (ISTFA) resulted in 13 papers from 2015 to 2022. A search of International Symposium on the Physical and Failure Analysis of Integrated Circuits (IPFA) resulted in 27 papers from 2018 to 2022. These papers were included in AI sections of the conferences or included “artificial intelligence” or “machine learning” in their titles. Regarding the academic databases, we searched papers also in IEEE Xplore and Google Scholar for each of the main categories of Symbolic and Sub-symbolic AI. For that, each name of the following categories was queried along with “failure analysis” or “semiconductor failure analysis:” General AI: • learning (artificial intelligence) AND failure analysis • artificial intelligence (or AI) AND failure analysis Symbolic: • logic AND failure analysis • logic inference AND failure analysis • expert systems AND failure analysis • query handler AND failure analysis • ontology AND failure analysis Sub-symbolic: Afterward, a manual evaluation of the publications was required, classifying the approaches stated by FA methods applied, AI categories applied and AI task (e.g. classification), input datatype and finally scope in the FA fault prediction.
RkJQdWJsaXNoZXIy MTMyMzg5NA==