Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Mohammad Hasan Ahmadilivani has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Can Model-Level Fault Tolerance be Enough for DNN Hardware Accelerators? A Study

    by Natalia Cherezova, Mohammad Hasan Ahmadilivani, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin

    As the hardware reliability of Deep Learning (DL) systems has become paramount due to extreme transistor scaling, designing a cost-efficient, reliable system is a major concern. Software-level strategies to improve the reliability of DL systems lack efficiency since they are agnostic to the underlying hardware. This work studies the effectiveness of the model-level reliability assessment and enhancement for DNNs at the software level for hardware accelerators and presents an end-to-end open-source framework ML-FRAMES to enable such an analysis. The paper addresses a critical question of whether or not model-level approaches correlate with hardware-level analyses and what the efficiency side effects are.


  • Open Access

    Progressive Training of Hybrid Clipped Rectified Linear Unit for Resilient Convolution Neural Networks

    by Seyedhamidreza Mousavi, Mohammad Hasan Ahmadilivani, Jaan Raik, Maksim Jenihhin, Masoud Daneshtalab

    Convolutional neural networks used in safety-critical systems must remain reliable under hardware-induced faults. Activation clipping is an effective fault-mitigation technique, yet existing approaches face a fundamental trade-off: layer-wise clipping provides limited resilience, whereas neuron-wise clipping introduces substantial memory overhead. This study proposes a hybrid clipped Rectified Linear Unit activation function that combines both strategies by applying neuron-wise clipping exclusively to the final layer and layer-wise clipping elsewhere. To further improve reliability, we introduce a progressive training strategy that optimizes the clipping thresholds in a layer-by-layer manner using knowledge distillation. This progressive optimization enables tighter threshold selection at each layer, improving fault tolerance while maintaining low memory costs. Experimental results under high error rates show that the proposed approach significantly enhances reliability compared with existing methods, achieving higher accuracy while substantially reducing memory overhead.

  • Open Access

    Real-Time Gait Anomaly Detection Using 1D-CNN and LSTM

    by Jakob Rostovski, Mohammad Hasan Ahmadilivani, Andrei Krivošei, Alar Kuusik, Muhammad Mahtab Alam

    Anomaly detection and fall prevention represent one of the key research areas within gait analysis for patients suffering from neurological disorders. Deep Learning has penetrated into healthcare applications, encompassing disease diagnosis and anomaly prediction. Connected wearable medical sensors are emerging due to computationally expensive machine learning tasks, which traditionally require use of remote PC or cloud computing. However, to reduce needs for wireless communication channel throughput, for data processing latency, and increase service reliability and safety, on device machine learning is gaining attention. This paper presents an innovative approach that leverages one dimensional convolutional neural network (1D-CNN) and long-short term memory (LSTM) neural network for the real-time detection of abnormal gait patterns during the step. Real-time anomaly detection pertains to the algorithm’s ability to promptly detect true gait abnormality occurrence during the swing phase of an ongoing step.For the experiments, we have collected eight different common gait anomalies, simulated by 22 persons, using motion sensors containing multidimensional inertial measurement units (IMUs).Results have demonstrated that the proposed 1D-CNN-AD algorithm achieves an average accuracy of 95% and an average F1-score of 88% for all gait types and can run in true real-time. Average earliness for 1D-CNN-AD algorithm was 0.6 s, which is mid-swing phase of the step. Proposed LSTM-AD algorithm achieved average accuracy of 87% and average F1-score of 70% for all gait types.


  • Top Cited Papers
    Open Access

    A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks

    by Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin

    Artificial Intelligence (AI) and, in particular, Machine Learning (ML), have emerged to be utilized in various applications due to their capability to learn how to solve complex problems. Over the past decade, rapid advances in ML have presented Deep Neural Networks (DNNs) consisting of a large number of neurons and layers. DNN Hardware Accelerators (DHAs) are leveraged to deploy DNNs in the target applications. Safety-critical applications, where hardware faults/errors would result in catastrophic consequences, also benefit from DHAs. Therefore, the reliability of DNNs is an essential subject of research. In recent years, several studies have been published accordingly to assess the reliability of DNNs. In this regard, various reliability assessment methods have been proposed on a variety of platforms and applications. Hence, there is a need to summarize the state-of-the-art to identify the gaps in the study of the reliability of DNNs. In this work, we conduct a Systematic Literature Review (SLR) on the reliability assessment methods of DNNs to collect relevant research works as much as possible, present a categorization of them, and address the open challenges. Through this SLR, three kinds of methods for reliability assessment of DNNs are identified, including Fault Injection (FI), Analytical, and Hybrid methods. Since the majority of works assess the DNN reliability by FI, we characterize different approaches and platforms of the FI method comprehensively. Moreover, Analytical and Hybrid methods are propounded. Thus, different reliability assessment methods for DNNs have been elaborated on their conducted DNN platforms and reliability evaluation metrics. Finally, we highlight the advantages and disadvantages of the identified methods and address the open challenges in the research area. We have concluded that Analytical and Hybrid methods are light-weight yet sufficiently accurate and have the potential to be extended in future research and to be utilized in establishing novel DNN reliability assessment frameworks.


  • ECS an endeavor towards providing similar cache reliability behavior in different programs

    by Mohammad Hasan Ahmadilivani, Mohammad Moeini Jahromi, Mostafa E. Salehi, Mona Kargar


  • Open Access

    A Systematic Literature Review on Hardware Reliability Assessment Methods for Deep Neural Networks

    by Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin

    Artificial Intelligence (AI) and, in particular, Machine Learning (ML) have emerged to be utilized in various applications due to their capability to learn how to solve complex problems. Over the last decade, rapid advances in ML have presented Deep Neural Networks (DNNs) consisting of a large number of neurons and layers. DNN Hardware Accelerators (DHAs) are leveraged to deploy DNNs in the target applications. Safety-critical applications, where hardware faults/errors would result in catastrophic consequences, also benefit from DHAs. Therefore, the reliability of DNNs is an essential subject of research. In recent years, several studies have been published accordingly to assess the reliability of DNNs. In this regard, various reliability assessment methods have been proposed on a variety of platforms and applications. Hence, there is a need to summarize the state of the art to identify the gaps in the study of the reliability of DNNs. In this work, we conduct a Systematic Literature Review (SLR) on the reliability assessment methods of DNNs to collect relevant research works as much as possible, present a categorization of them, and address the open challenges. Through this SLR, three kinds of methods for reliability assessment of DNNs are identified including Fault Injection (FI), Analytical, and Hybrid methods. Since the majority of works assess the DNN reliability by FI, we characterize different approaches and platforms of the FI method comprehensively. Moreover, Analytical and Hybrid methods are propounded. Thus, different reliability assessment methods for DNNs have been elaborated on their conducted DNN platforms and reliability evaluation metrics. Finally, we highlight the advantages and disadvantages of the identified methods and address the open challenges in the research area.

  • Open Access

    DeepVigor: Vulnerability Value Ranges and Factors for DNNs' Reliability Assessment

    by Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin

    Deep Neural Networks (DNNs) and their accelerators are being deployed ever more frequently in safety-critical applications leading to increasing reliability concerns. A traditional and accurate method for assessing DNNs' reliability has been resorting to fault injection, which, however, suffers from prohibitive time complexity. While analytical and hybrid fault injection-/analytical-based methods have been proposed, they are either inaccurate or specific to particular accelerator architectures. In this work, we propose a novel accurate, fine-grain, metric-oriented, and accelerator-agnostic method called DeepVigor that provides vulnerability value ranges for DNN neurons' outputs. An outcome of DeepVigor is an analytical model representing vulnerable and non-vulnerable ranges for each neuron that can be exploited to develop different techniques for improving DNNs' reliability. Moreover, DeepVigor provides reliability assessment metrics based on vulnerability factors for bits, neurons, and layers using the vulnerability ranges. The proposed method is not only faster than fault injection but also provides extensive and accurate information about the reliability of DNNs, independent from the accelerator. The experimental evaluations in the paper indicate that the proposed vulnerability ranges are 99.9% to 100% accurate even when evaluated on previously unseen test data. Also, it is shown that the obtained vulnerability factors represent the criticality of bits, neurons, and layers proficiently. DeepVigor is implemented in the PyTorch framework and validated on complex DNN benchmarks.

  • Open Access

    ECS an Endeavor Towards Providing Similar Cache Reliability Behavior in Different Programs

    by Mohammad Hasan Ahmadilivani, Mohammad Moeini Jahromi, Mostafa E. Salehi, Mona Kargar

    The reliability of embedded processors is one of the major concerns in safety-critical applications. Reliability is particularly expressed within the cache memories which are the largest part of new system on chips. Cache memories are the most vulnerable parts of the embedded systems and can affect the reliability drastically especially in deep transistor scaling. Therefore, evaluating the cache vulnerability is crucial in the design of a reliable system especially for safety-critical applications. It has been shown that using the same cache sizes for different programs leads to incompatible vulnerability patterns in them. According to the literature, most of the related researches, have exploited identical cache sizes for different programs in their reliability evaluations, while the cache reliability strictly depends on the cache size and program behavior. Traditional attempts for finding an appropriate cache size for different programs would need a huge design space exploration. In this work, we have introduced a criterion for determining the Effective Cache Size (ECS) for embedded processors which considers the inherent programs’ reliability and performance properties. According to the results, using the ECS for the representative benchmark applications, the reliability would be increased 43x on average with acceptable performance degradations (21% on average).