Dr. Mahdi Taheri
Mahdi Taheri received his B.Sc. degree in Electrical Engineering in 2017 and his M.Sc. degree in Electrical Engineering in 2021. He obtained his Ph.D. in Computer Systems from Tallinn University of Technology (TalTech), Estonia, in January 2025. During his doctoral studies, he gained industry-oriented research experience through a six-month internship at IHP – Leibniz Institute for High Performance Microelectronics, Germany. From January 2025, he has been working as a researcher at Brandenburg University of Technology (BTU) Cottbus-Senftenberg, Germany, while continuing as a part-time researcher at TalTech. He joined Humboldt University of Berlin as a permanent researcher in September 2026 and leads his Reliable and Secure Systems (RSS) team, with research focused on reliable and secure AI systems, neuromorphic computing, approximate computing, and hardware accelerators.
Thesis
Thesis and dissertation materials
Hosted thesis information attached to this public profile.
Methods for Reliability Assessment and Enhancement of Deep Neural Networks Hardware Accelerators
Research Topics
Related topics
Research areas currently linked to this public profile.
Projects
Related projects
Public projects currently associated with this profile.
Publications
Linked publications
Selected publications currently associated with this profile.
2026
- 20 July 20260 citations
PRISM: Sensitivity-Aware PolynoMial PRuning for EffIcient Neural Network Encryption
S Majavdia, M Taheri
arXiv preprint arXiv:2607.18342, 2026
- 1 July 20260 citations
Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers
A Pradeep, S Nazari, M Taheri, C Herglotz
arXiv preprint arXiv:2607.02612, 2026
- 30 May 20260 citations
HASA: Subnet Allocation for Compute-Constrained Model-Heterogeneous Federated Learning
AH Shahdadian, AM Abdelmoniem, M Taheri, S Nazari, C Herglotz
arXiv preprint arXiv:2606.07621, 2026
- 30 May 20260 citations
SENTRY: Statistical Reliability Analysis of Vision Transformers Under Soft Errors
PK Bhaduri, M Taheri, S Nazari, M Jenihhin, C Herglotz, M Hubner
arXiv preprint arXiv:2606.07620, 2026
- 24 May 20261 citations
An FPGA-Based SoC Architecture with a RISC-V Controller for Energy-Efficient Temporal-Coding Spiking Neural Networks
MJ Sekonji, A Mahani, M Mirsadeghi, M Taheri
2026 IEEE International Symposium on Circuits and Systems (ISCAS), 4573-4577, 2026
- 23 May 20260 citations
Quantization-Aided Cost-Efficient Reliability of CNN Accelerators for Edge AI
M Jenihhin, M Taheri, N Cherezova
Machine Learning Systems: The Role of Hardware Design for Dependable …, 2026
- 20 April 20264 citations
HAWX: A hardware-aware framework for fast and scalable approximation of DNNs
S Nazari, MS Almasi, M Taheri, A Azarpeyvand, A Mokhtari, A Mahani, ...
2026 Design, Automation & Test in Europe Conference (DATE), 1-7, 2026
- 17 March 20263 citations
Mix-and-match pruning: Globally guided layer-wise sparsification of dnns
D Monachan, S Nazari, M Taheri, A Azarpeyvand, M Krstic, M Huebner, ...
arXiv preprint arXiv:2603.20280, 2026
- 17 March 20261 citations
RESQ: A Unified Framework for ReLiability-and SEcurity Enhancement of Quantized Deep Neural Networks
AS Mohammadi, S Nazari, A Azarpeyvand, M Taheri, M Krstic, M Hübner, ...
2026 IEEE 27th Latin American Test Symposium (LATS), 1-4, 2026
- 15 March 20261 citations
SPARQ: Spiking Early-Exit Neural Networks for Energy-Efficient Edge AI
P Patne, M Taheri, A Mahani, M Jenihhin, R Mahani, C Herglotz
arXiv preprint arXiv:2603.14380, 2026
- 24 February 20262 citations
Sensitivity-Guided Framework for Pruned and Quantized Reservoir Computing Accelerators
A Jafari, M Taheri, HG Mohammadi, C Herglotz, M Platzner
arXiv preprint arXiv:2603.08737, 2026
- 23 February 20260 citations
Integrating an open-source soft-GPU overlay with RISC-V control and high-bandwidth memory
HGM Hernandez, M Taheri, M Ali, K Shahin, A Syavashi, D Göhringer, ...
Journal of Systems Architecture, 103736, 2026
- 17 February 20260 citations
DART: Input-Difficulty-AwaRe Adaptive Threshold for Early-Exit DNNs
P Patne, M Taheri, C Herglotz, M Jenihhin, M Krstic, M Hübner
arXiv preprint arXiv:2603.12269, 2026
- 17 February 20261 citations
PhD Thesis Summary: Methods for Reliability Assessment and Enhancement of Deep Neural Network Hardware Accelerators
M Taheri
arXiv preprint arXiv:2603.08724, 2026
- 20260 citations
Integrating an open-source soft-GPU overlay with RISC-V control and high-bandwidth memory
HG Muñoz Hernandez, M Taheri, M Ali, K Shahin, A Syavashi, ...
2025
- 16 December 20253 citations
Adaptive fault resilience for early-exit dnns
RM Kodamanchili, N Cherezova, M Taheri, M Jenihhin
2025 IEEE International Test Conference in Asia (ITC-Asia), 108-113, 2025
- 16 December 20250 citations
SAFFIRA A Framework for Assessing the Reliability of Systolic-Array DNN Accelerators
S Pappalardo, N Bellarmino, B Deveautour, A Bosio, M Taheri, ...
Journal of Circuits, Systems and Computers 34 (18), 2543001, 2025
- 2 September 20251 citations
Reliability-aware hyperparameter optimization for ANN-to-SNN conversion
S Sharifian, M Taheri, V Rashtchi, A Azarpeyvand, C Herglotz, M Jenihhin
WiPiEC Journal-Works in Progress in Embedded Computing Journal 11 (1), 7-7, 2025
- 7 July 20251 citations
Shield: Pso-based hardware trojan detection for efficient and low-cost defense
M Hosseini, A Azarpeyvand, M Taheri, T Ghasempouri, M Jenihhin
2025 IEEE 31st International Symposium on On-Line Testing and Robust System …, 2025
- 5 May 20255 citations
Rl-agent-based early-exit dnn architecture search framework
M Taheri, P Patne, N Cherezova, A Mahani, C Herglotz, M Jenihhin
2025 IEEE 28th International Symposium on Design and Diagnostics of …, 2025
- 28 March 20255 citations
Genie: Genetic algorithm-based reliability assessment methodology for deep neural networks
S Nazari, M Taheri, A Azarpeyvand, M Afsharchi, C Herglotz, M Jenihhin
2025 11th International Conference on Computing and Artificial Intelligence …, 2025
- 11 March 20255 citations
Reliability-aware performance optimization of dnn hw accelerators through heterogeneous quantization
S Nazari, M Taheri, A Azarpeyvand, M Afsharchi, C Herglotz, M Jenihhin
2025 IEEE 26th Latin American Test Symposium (LATS), 1-6, 2025
2024
- 27 December 202442 citations
AdAM: Adaptive Approximate Multiplier for Fault Tolerance in DNN Accelerators
M Taheri, N Cherezova, S Nazari, A Azarpeyvand, T Ghasempouri, ...
IEEE Transactions on Device and Materials Reliability, 2024
- 6 October 20241 citations
Heterogeneous approximation of dnn hw accelerators based on channels vulnerability
N Cherezova, S Pappalardo, M Taheri, MH Ahmadilivani, B Deveautour, ...
2024 IFIP/IEEE 32nd International Conference on Very Large Scale Integration …, 2024
- 22 April 202411 citations
Special session: Reliability assessment recipes for dnn accelerators
MH Ahmadilivani, A Bosio, B Deveautour, FF Dos Santos, ...
2024 IEEE 42nd VLSI Test Symposium (VTS), 1-11, 2024
- 9 April 20240 citations
Keynote: Cost-Efficient Reliability for Edge-AI Chips
M Jenihhin, M Taheri, N Cherezova, MH Ahmadilivani, H Selg, A Jutman, ...
2024 IEEE 25th Latin American Test Symposium (LATS), 1-2, 2024
- 3 April 202432 citations
Exploration of activation fault reliability in quantized systolic array-based dnn accelerators
M Taheri, N Cherezova, MS Ansari, M Jenihhin, A Mahani, ...
2024 25th International Symposium on Quality Electronic Design (ISQED), 1-8, 2024
- 3 April 202429 citations
Saffira: a framework for assessing the reliability of systolic-array-based dnn accelerators
M Taheri, M Daneshtalab, J Raik, M Jenihhin, S Pappalardo, P Jimenez, ...
2024 27th International Symposium on Design & Diagnostics of Electronic …, 2024
- 22 January 2024126 citations
A systematic literature review on hardware reliability assessment methods for deep neural networks
MH Ahmadilivani, M Taheri, J Raik, M Daneshtalab, M Jenihhin
ACM Computing Surveys 56 (6), 1-39, 2024
- 20248 citations
FORTUNE: A Negative Memory Overhead Hardware-Agnostic Fault TOleRance TechniqUe in DNNs
S Nazari, M Taheri, A Azarpeyvand, M Afsharchi, T Ghasempouri, ...
Authorea Preprints, 2024
2023
- 11 June 202321 citations
Enhancing fault resilience of qnns by selective neuron splitting
MH Ahmadilivani, M Taheri, J Raik, M Daneshtalab, M Jenihhin
2023 IEEE 5th International Conference on Artificial Intelligence Circuits …, 2023
- 22 May 202324 citations
Deepvigor: Vulnerability value ranges and factors for dnns’ reliability assessment
MH Ahmadilivani, M Taheri, J Raik, M Daneshtalab, M Jenihhin
2023 IEEE European Test Symposium (ETS), 1-6, 2023
- 3 May 202317 citations
Appraiser: Dnn fault resilience analysis employing approximation errors
M Taheri, MH Ahmadilivani, M Jenihhin, M Daneshtalab, J Raik
2023 26th International Symposium on Design and Diagnostics of Electronic …, 2023
- 26 April 20233 citations
Noise-tolerance gpu-based age estimation using resnet-50
M Taheri, M Taheri, A Hadjahmadi
arXiv preprint arXiv:2305.00848, 2023
- 24 April 202335 citations
Special session: Approximation and fault resiliency of dnn accelerators
MH Ahmadilivani, M Barbareschi, S Barone, A Bosio, M Daneshtalab, ...
2023 IEEE 41st VLSI Test Symposium (VTS), 1-10, 2023
- 5 April 202336 citations
Deepaxe: A framework for exploration of approximation and reliability trade-offs in dnn accelerators
M Taheri, M Riazati, MH Ahmadilivani, M Jenihhin, M Daneshtalab, J Raik, ...
2023 24th International Symposium on Quality Electronic Design (ISQED), 1-8, 2023
- 15 March 20231 citations
LRDB: LSTM raw data dna base-caller based on long-short term models in an active learning environment
A Rezaei, M Taheri, A Mahani, S Magierowski
arXiv preprint arXiv:2303.08915, 2023
2022
- 12 September 20224 citations
A novel fault-tolerant logic style with self-checking capability
M Taheri, S Sheikhpour, A Mahani, M Jenihhin
2022 IEEE 28th International Symposium on On-Line Testing and Robust System …, 2022
- May 202213 citations
Dnn hardware reliability assessment and enhancement
M Taheri
27th IEEE European Test Symposium (ETS), 2022
- 1 January 20222 citations
A fault-resistant architecture for aes s-box architecture
M Taheri, S Sheikhpour, MS Ansari, A Mahani
Journal of Applied Research in Electrical Engineering 1 (1), 86-92, 2022
2021
- 15 August 20211 citations
A Novel 2-D BWA-MEM FPGA Accelerator for Short-Read Mapping of the Whole Human Genome
M Taheri, A Mahani
Journal of Applied Research in Electrical Engineering 1 (2), 203-210, 2021
- 27 May 20215 citations
A high-performance MEMRISTOR-based Smith-Waterman DNA sequence alignment Using FPNI structure
M Taheri, H Zandevakili, A Mahani
Journal of Applied Research in Electrical Engineering 1 (1), 2021
- 17 May 20214 citations
Strengthened 32‐bit AES implementation: Architectural error correction configuration with a new voting scheme
S Sheikhpur, M Taheri, MS Ansari, A Mahani
IET Computers & Digital Techniques 15 (3), 1-14, 2021
- 23 February 20212 citations
Hardware acceleration of the novel two dimensional Burrows‐Wheeler Aligner algorithm with maximal exact matches seed extension kernel
M Taheri, MS Ansari, S Magierowski, A Mahani
IET Circuits, Devices & Systems 15 (2), 94-103, 2021
- 22 February 20214 citations
Development and hardware acceleration of a novel 2-DBWA-MEM DNA sequencing alignment algorithm
M Taheri, A Mahani
- 22 February 20215 citations
DMR-based Technique for Fault Tolerant AES S-box Architecture
M Taheri, S Sheikhpour, MS Ansari, A Mahani
1 st Conference on Applied Research in Electrical Engineering (AREE), 2021
Year pending
- 0 citations
28th IEEE International Symposium on Design and Diagnostics of Electronic Circuits and Systems-DDECS 2025
H Abdo, J Lappas, M Esmaeilpour, C Weis, N Wehn, AH Hadipour, ...
- 0 citations
SPECIAL SECTION ON DFT SYMPOSIUM
D Mateo, X Aragones, E Barajas, SM Domingo, X Gisbert, J Altet, ...
