New techniques for the Reliability Evaluation of AI-oriented Hardware Accelerators

• NA (2025)

Master of Science

• GPU programming. A.A. 2025/26, INGEGNERIA INFORMATICA (COMPUTER ENGINEERING). Collaboratore del corso

Bachelor of Science

• Computer sciences. A.A. 2023/24, INGEGNERIA DELL'AUTOVEICOLO (AUTOMOTIVE ENGINEERING). Collaboratore del corso
• Computer sciences. A.A. 2024/25, INGEGNERIA DELL'AUTOVEICOLO (AUTOMOTIVE ENGINEERING). Collaboratore del corso

• CAD - Electronic CAD & Reliability Group

• Ahmadilivani, Mohammad Hasan; Cherezova, Natalia; Glaß, Michael; Guerrero-Balaguera, ... (2026)
  Reliability Assessment of Deep Neural Networks and Accelerators Across Design Stages. In: 27th Latin American Test Symposium, LATS 2026, Florianópolis (BRA), 17-20 March 2026, pp. 1-10. ISBN: 979-8-3195-4235-9
  Contributo in Atti di Convegno (Proceeding)
• Limas Sierra, Robert; Guerrero-Balaguera, Juan-David; Rodriguez Condia, Josie Esteban; ... (2026)
  A Software-based Fault Tolerance Mechanism for Matrix Multiplication Operations in Tensor Cores. In: IEEE TRANSACTIONS ON COMPUTERS, pp. 1-14. ISSN 0018-9340
  Contributo su Rivista
• Limas Sierra, Robert Alexander (2026)
  New techniques for the Reliability Evaluation of AI-oriented Hardware Accelerators. relatore: SONZA REORDA, MATTEO; RODRIGUEZ CONDIA, JOSIE ESTEBAN; , 38. XXXVIII Ciclo, P.: 109
  Doctoral Thesis
• Guerrero-Balaguera, Juan-David; Limas Sierra, Robert; Vilar De Farias, Gustavo; Abed, ... (2026)
  Analyzing hardware accelerators’ reliability: from design exploration to fast evaluation with hyperscalers. In: 2026 17th IEEE Latin American Symposium on Circuits and Systems, Arequipa (PE), 24-27 February 2026. ISBN: 979-8-3315-7097-2
  Contributo in Atti di Convegno (Proceeding)
• Rodriguez Condia, Josie Esteban; Guerrero-Balaguera, Juan-David; Sierra, Robert Limas; ... (2025)
  Investigating and Mitigating Critical Faults in Floating-Point and Posit Arithmetic Hardware. In: IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, vol. 13, pp. 1605-1617. ISSN 2168-6750
  Contributo su Rivista
• Guerrero-Balaguera, Juan-David; Rodriguez condia, Josie Esteban; Sierra, Robert Limas; ... (2025)
  About the Functional In-Field Self-Testing of AI Accelerators. In: 2025 IEEE 26th Latin American Test Symposium (LATS), San Andres Island (COL), 11-14 March 2025, pp. 1-6. ISBN: 978-1-6654-7763-5
  Contributo in Atti di Convegno (Proceeding)
• Caro-Anzola, Edward-W.; Esposito, Giuseppe; Guerrero-Balaguera, Juan-David; ... (2025)
  IoT and Edge Computing: Applications and Reliability Implications. In: 2025 IEEE 26th Latin American Test Symposium (LATS), San Andres Island (COL), 11-14 March 2025, pp. 1-10. ISBN: 978-1-6654-7763-5
  Contributo in Atti di Convegno (Proceeding)
• Limas Sierra, Robert; Guerrero Balaguera, Juan David; Rodriguez Condia, Josie E.; Sonza ... (2025)
  A Structured Method to Generate Self-Test Libraries for Tensor Cores. In: ELECTRONICS, vol. 14. ISSN 2079-9292
  Contributo su Rivista
• Sierra, Robert Limas; Esposito, Giuseppe; Guerrero-Balaguera, Juan-David; Rodriguez ... (2025)
  Improving CNN Runtime Robustness Against Soft Errors by Dropout Layer Optimization. In: 2025 IEEE 26th Latin American Test Symposium (LATS), San Andres Islas (COL), 11-14 March 2025, pp. 1-6. ISBN: 978-1-6654-7763-5
  Contributo in Atti di Convegno (Proceeding)
• Guerrero-Balaguera, Juan-David; Limas-Sierra, Robert; Sensoz, Oguz; Rodriguez Condia, ... (2025)
  SHADOWFI: An Open-source Framework For Fault Evaluation of Complex IC Designs Using Hyperscale Computing. In: IEEE ACCESS, vol. 13, pp. 211382-211406. ISSN 2169-3536
  Contributo su Rivista
• Rodriguez Condia, Josie E.; Guerrero-Balaguera, Juan-David; Patiño Núñez, Edwar J.; ... (2024)
  Investigating and Reducing the Architectural Impact of Transient Faults in Special Function Units for GPUs. In: JOURNAL OF ELECTRONIC TESTING, vol. 40, pp. 215-228. ISSN 0923-8174
  Contributo su Rivista
• Limas Sierra, Robert; Guerrero-Balaguera, Juan-David; Rodriguez Condia, Josie E.; Sonza ... (2024)
  Exploring Hardware Fault Impacts on Different Real Number Representations of the Structural Resilience of TCUs in GPUs. In: ELECTRONICS, vol. 13. ISSN 2079-9292
  Contributo su Rivista
• Limas Sierra, Robert; Guerrero-Balaguera, Juan-David; Pessia, Francesco; Rodriguez ... (2024)
  Analyzing the Impact of Scheduling Policies on the Reliability of GPUs Running CNN Operations. In: 2024 IEEE 42nd VLSI Test Symposium (VTS), Tempe, AZ (USA), 22-24 April 2024. ISBN: 979-8-3503-6378-4
  Contributo in Atti di Convegno (Proceeding)
• Ahmadilivani, Mohammad Hasan; Bosio, Alberto; Deveautour, Bastien; Santos, Fernando ... (2024)
  Special Session: Reliability Assessment Recipes for DNN Accelerators. In: 2024 IEEE 42nd VLSI Test Symposium (VTS), Tempe (USA), 22-24 April 2024. ISBN: 979-8-3503-6378-4
  Contributo in Atti di Convegno (Proceeding)
• Limas Sierra, Robert Alexander; Guerrero-Balaguera, Juan-David; Rodriguez Condia, Josie ... (2024)
  Analyzing the Reliability of TCUs Through Micro-architecture and Structural Evaluations for Two Real Number Formats. In: 31st IFIP WG 10.5/IEEE International Conference on Very Large Scale Integration, VLSI-SoC 2023, Sharjah, United Arab Emirates, October 16–18, 2023, pp. 149-176. ISSN 1868-4238. ISBN: 978-3-031-70946-3
  Contributo in Atti di Convegno (Proceeding)
• Chen, Junchao; Esposito, Giuseppe; Fernandes dos Santos, Fernando; Guerrero-Balaguera, ... (2024)
  Reliability Assessment of Large DNN Models:Trading Off Performance and Accuracy. In: 2024 IFIP/IEEE 32nd International Conference on Very Large Scale Integration (VLSI-SoC), Tanger (MAR), 06-09 October 2024, pp. 1-10. ISBN: 979-8-3315-3967-2
  Contributo in Atti di Convegno (Proceeding)
• Rodriguez Condia, Josie E; Guerrero-Balaguera, Juan-David; Sierra, Robert Limas; Reorda, ... (2024)
  Analyzing the Structural and Operational Impact of Faults in Floating-Point and Posit Arithmetic Cores for CNN Operations. In: 29th IEEE European Test Symposium 2024, The Hague (NL), 20-24 May 2024, pp. 1-4. ISBN: 979-8-3503-4932-0
  Contributo in Atti di Convegno (Proceeding)
• Pessia, Francesco; Guerrero-Balaguera, Juan-David; Limas Sierra, Robert; Rodriguez ... (2024)
  Effective Application-level Error Modeling of Permanent Faults on AI Accelerators. In: International Symposium on On-Line Testing and Robust System Design (IOLTS), Rennes (FRA), 03-05 July 2024, pp. 1-7. ISBN: 979-8-3503-7055-3
  Contributo in Atti di Convegno (Proceeding)
• Sierra, Robert Limas; Guerrero-Balaguera, Juan-David; Rodriguez Condia, Josie E.; Sonza ... (2023)
  Analyzing the Impact of Different Real Number Formats on the Structural Reliability of TCUs in GPUs. In: IFIP/IEEE Conference on Very Large Scale Integration (VLSI-SoC 2023), Dubai (United Arab Emirates), 16-18 October 2023, pp. 1-6. ISBN: 979-8-3503-2599-7
  Contributo in Atti di Convegno (Proceeding)
• Rodriguez Condia, Josie E.; Guerrero-Balaguera, Juan-David; Patiño Núñez, Edwar J.; ... (2023)
  Evaluating the Prevalence of SFUs in the Reliability of GPUs. In: 2023 IEEE European Test Symposium (ETS), Venice (IT), 22-26 May 2023, pp. 1-6. ISBN: 979-8-3503-3634-4
  Contributo in Atti di Convegno (Proceeding)