Maksim Jenihhin

Mahdi Taheri, Mohammad Riazati, Mohammad Hasan Ahmadilivani et al.

While the role of Deep Neural Networks (DNNs) in a wide range of safety-critical applications is expanding, emerging DNNs experience massive growth in terms of computation power. It raises the necessity of improving the reliability of DNN accelerators yet reducing the computational burden on the hardware platforms, i.e. reducing the energy consumption and execution time as well as increasing the efficiency of DNN accelerators. Therefore, the trade-off between hardware performance, i.e. area, power and delay, and the reliability of the DNN accelerator implementation becomes critical and requires tools for analysis. In this paper, we propose a framework DeepAxe for design space exploration for FPGA-based implementation of DNNs by considering the trilateral impact of applying functional approximation on accuracy, reliability and hardware performance. The framework enables selective approximation of reliability-critical DNNs, providing a set of Pareto-optimal DNN implementation design space points for the target resource utilization requirements. The design flow starts with a pre-trained network in Keras, uses an innovative high-level synthesis environment DeepHLS and results in a set of Pareto-optimal design space points as a guide for the designer. The framework is demonstrated in a case study of custom and state-of-the-art DNNs and datasets.

Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik et al.

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.

Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik et al.

The superior performance of Deep Neural Networks (DNNs) has led to their application in various aspects of human life. Safety-critical applications are no exception and impose rigorous reliability requirements on DNNs. Quantized Neural Networks (QNNs) have emerged to tackle the complexity of DNN accelerators, however, they are more prone to reliability issues. In this paper, a recent analytical resilience assessment method is adapted for QNNs to identify critical neurons based on a Neuron Vulnerability Factor (NVF). Thereafter, a novel method for splitting the critical neurons is proposed that enables the design of a Lightweight Correction Unit (LCU) in the accelerator without redesigning its computational part. The method is validated by experiments on different QNNs and datasets. The results demonstrate that the proposed method for correcting the faults has a twice smaller overhead than a selective Triple Modular Redundancy (TMR) while achieving a similar level of fault resiliency.

Tanvir Ahmad Tarique, Foisal Ahmed, Maksim Jenihhin et al.

Recently, recycled field-programmable gate arrays (FPGAs) pose a significant hardware security problem due to the proliferation of the semiconductor supply chain. Ring oscillator (RO) based frequency analyzing technique is one of the popular methods, where most studies used the known fresh FPGAs (KFFs) in machine learning-based detection, which is not a realistic approach. In this paper, we present a novel recycled FPGA detection method by examining the symmetry information of the RO frequency using unsupervised anomaly detection method. Due to the symmetrical array structure of the FPGA, some adjacent logic blocks on an FPGA have comparable RO frequencies, hence our method simply analyzes the RO frequencies of those blocks to determine how similar they are. The proposed approach efficiently categorizes recycled FPGAs by utilizing direct density ratio estimation through outliers detection. Experiments using Xilinx Artix-7 FPGAs demonstrate that the proposed method accurately classifies recycled FPGAs from 10 fresh FPGAs by x fewer computations compared with the conventional method.

Parth Patne, Mahdi Taheri, Ali Mahani et al.

Spiking neural networks (SNNs) offer inherent energy efficiency due to their event-driven computation model, making them promising for edge AI deployment. However, their practical adoption is limited by the computational overhead of deep architectures and the absence of input-adaptive control. This work presents SPARQ, a unified framework that integrates spiking computation, quantization-aware training, and reinforcement learning-guided early exits for efficient and adaptive inference. Evaluations across MLP, LeNet, and AlexNet architectures demonstrated that the proposed Quantised Dynamic SNNs (QDSNN) consistently outperform conventional SNNs and QSNNs, achieving up to 5.15% higher accuracy over QSNNs, over 330 times lower system energy compared to baseline SNNs, and over 90 percent fewer synaptic operations across different datasets. These results validate SPARQ as a hardware-friendly, energy-efficient solution for real-time AI at the edge.

Mahdi Taheri, Natalia Cherezova, Mohammad Saeed Ansari et al.

The stringent requirements for the Deep Neural Networks (DNNs) accelerator's reliability stand along with the need for reducing the computational burden on the hardware platforms, i.e. reducing the energy consumption and execution time as well as increasing the efficiency of DNN accelerators. Moreover, the growing demand for specialized DNN accelerators with tailored requirements, particularly for safety-critical applications, necessitates a comprehensive design space exploration to enable the development of efficient and robust accelerators that meet those requirements. Therefore, the trade-off between hardware performance, i.e. area and delay, and the reliability of the DNN accelerator implementation becomes critical and requires tools for analysis. This paper presents a comprehensive methodology for exploring and enabling a holistic assessment of the trilateral impact of quantization on model accuracy, activation fault reliability, and hardware efficiency. A fully automated framework is introduced that is capable of applying various quantization-aware techniques, fault injection, and hardware implementation, thus enabling the measurement of hardware parameters. Moreover, this paper proposes a novel lightweight protection technique integrated within the framework to ensure the dependable deployment of the final systolic-array-based FPGA implementation. The experiments on established benchmarks demonstrate the analysis flow and the profound implications of quantization on reliability, hardware performance, and network accuracy, particularly concerning the transient faults in the network's activations.

Mahdi Taheri, Natalia Cherezova, Samira Nazari et al.

In this paper, we propose an architecture of a novel adaptive fault-tolerant approximate multiplier tailored for ASIC-based DNN accelerators.

Mahdi Taheri, Masoud Daneshtalab, Jaan Raik et al.

Systolic array has emerged as a prominent architecture for Deep Neural Network (DNN) hardware accelerators, providing high-throughput and low-latency performance essential for deploying DNNs across diverse applications. However, when used in safety-critical applications, reliability assessment is mandatory to guarantee the correct behavior of DNN accelerators. While fault injection stands out as a well-established practical and robust method for reliability assessment, it is still a very time-consuming process. This paper addresses the time efficiency issue by introducing a novel hierarchical software-based hardware-aware fault injection strategy tailored for systolic array-based DNN accelerators.

Mohammad Hasan Ahmadilivani, Seyedhamidreza Mousavi, Jaan Raik et al.

Convolutional Neural Networks (CNNs) have become integral in safety-critical applications, thus raising concerns about their fault tolerance. Conventional hardware-dependent fault tolerance methods, such as Triple Modular Redundancy (TMR), are computationally expensive, imposing a remarkable overhead on CNNs. Whereas fault tolerance techniques can be applied either at the hardware level or at the model levels, the latter provides more flexibility without sacrificing generality. This paper introduces a model-level hardening approach for CNNs by integrating error correction directly into the neural networks. The approach is hardware-agnostic and does not require any changes to the underlying accelerator device. Analyzing the vulnerability of parameters enables the duplication of selective filters/neurons so that their output channels are effectively corrected with an efficient and robust correction layer. The proposed method demonstrates fault resilience nearly equivalent to TMR-based correction but with significantly reduced overhead. Nevertheless, there exists an inherent overhead to the baseline CNNs. To tackle this issue, a cost-effective parameter vulnerability based pruning technique is proposed that outperforms the conventional pruning method, yielding smaller networks with a negligible accuracy loss. Remarkably, the hardened pruned CNNs perform up to 24\% faster than the hardened un-pruned ones.

Natalia Cherezova, Artur Jutman, Maksim Jenihhin

The emergence of Deep Neural Networks (DNNs) in mission- and safety-critical applications brings their reliability to the front. High performance demands of DNNs require the use of specialized hardware accelerators. Systolic array architecture is widely used in DNN accelerators due to its parallelism and regular structure. This work presents a run-time reconfigurable systolic array architecture with three execution modes and four implementation options. All four implementations are evaluated in terms of resource utilization, throughput, and fault tolerance improvement. The proposed architecture is used for reliability enhancement of DNN inference on systolic array through heterogeneous mapping of different network layers to different execution modes. The approach is supported by a novel reliability assessment method based on fault propagation analysis. It is used for the exploration of the appropriate execution mode--layer mapping for DNN inference. The proposed architecture efficiently protects registers and MAC units of systolic array PEs from transient and permanent faults. The reconfigurability feature enables a speedup of up to $3\times$, depending on layer vulnerability. Furthermore, it requires $6\times$ fewer resources compared to static redundancy and $2.5\times$ fewer resources compared to the previously proposed solution for transient faults.

Parth Patne, Mahdi Taheri, Christian Herglotz et al.

Early-exit deep neural networks enable adaptive inference by terminating computation when sufficient confidence is achieved, reducing cost for edge AI accelerators in resource-constrained settings. Existing methods, however, rely on suboptimal exit policies, ignore input difficulty, and optimize thresholds independently. This paper introduces DART (Input-Difficulty-Aware Adaptive Threshold), a framework that overcomes these limitations. DART introduces three key innovations: (1) a lightweight difficulty estimation module that quantifies input complexity with minimal computational overhead, (2) a joint exit policy optimization algorithm based on dynamic programming, and (3) an adaptive coefficient management system. Experiments on diverse DNN benchmarks (AlexNet, ResNet-18, VGG-16) demonstrate that DART achieves up to \textbf{3.3$\times$} speedup, \textbf{5.1$\times$} lower energy, and up to \textbf{42\%} lower average power compared to static networks, while preserving competitive accuracy. Extending DART to Vision Transformers (LeViT) yields power (5.0$\times$) and execution-time (3.6$\times$) gains but also accuracy loss (up to 17 percent), underscoring the need for transformer-specific early-exit mechanisms. We further introduce the Difficulty-Aware Efficiency Score (DAES), a novel multi-objective metric, under which DART achieves up to a 14.8 improvement over baselines, highlighting superior accuracy, efficiency, and robustness trade-offs.

Mohammad Hasan Ahmadilivani, Jaan Raik, Masoud Daneshtalab et al.

The growing exploitation of Machine Learning (ML) in safety-critical applications necessitates rigorous safety analysis. Hardware reliability assessment is a major concern with respect to measuring the level of safety in ML-based systems. Quantifying the reliability of emerging ML models, including Convolutional Neural Networks (CNNs), is highly complex due to their enormous size in terms of the number of parameters and computations. Conventionally, Fault Injection (FI) is applied to perform a reliability measurement. However, performing FI on modern-day CNNs is prohibitively time-consuming if an acceptable confidence level is to be achieved. To speed up FI for large CNNs, statistical FI (SFI) has been proposed, but its runtimes are still considerably long. In this work, we introduce DeepVigor+, a scalable, fast, and accurate semi-analytical method as an efficient alternative for reliability measurement in CNNs. DeepVigor+ implements a fault propagation analysis model and attempts to acquire Vulnerability Factors (VFs) as reliability metrics in an optimal way. The results indicate that DeepVigor+ obtains VFs for CNN models with an error less than $1\%$, i.e., the objective in SFI, but with $14.9$ up to $26.9$ times fewer simulations than the best-known state-of-the-art SFI. DeepVigor+ enables an accurate reliability analysis for large and deep CNNs within a few minutes, rather than achieving the same results in days or weeks.

Seyedhamidreza Mousavi, Mohammad Hasan Ahmadilivani, Jaan Raik et al.

Deep Neural Networks (DNNs) are extensively employed in safety-critical applications where ensuring hardware reliability is a primary concern. To enhance the reliability of DNNs against hardware faults, activation restriction techniques significantly mitigate the fault effects at the DNN structure level, irrespective of accelerator architectures. State-of-the-art methods offer either neuron-wise or layer-wise clipping activation functions. They attempt to determine optimal clipping thresholds using heuristic and learning-based approaches. Layer-wise clipped activation functions cannot preserve DNNs resilience at high bit error rates. On the other hand, neuron-wise clipping activation functions introduce considerable memory overhead due to the addition of parameters, which increases their vulnerability to faults. Moreover, the heuristic-based optimization approach demands numerous fault injections during the search process, resulting in time-consuming threshold identification. On the other hand, learning-based techniques that train thresholds for entire layers concurrently often yield sub-optimal results. In this work, first, we demonstrate that it is not essential to incorporate neuron-wise activation functions throughout all layers in DNNs. Then, we propose a hybrid clipped activation function that integrates neuron-wise and layer-wise methods that apply neuron-wise clipping only in the last layer of DNNs. Additionally, to attain optimal thresholds in the clipping activation function, we introduce ProAct, a progressive training methodology. This approach iteratively trains the thresholds on a layer-by-layer basis, aiming to obtain optimal threshold values in each layer separately.

Mohammad Hasan Ahmadilivani, Mario Barbareschi, Salvatore Barone et al.

Deep Learning, and in particular, Deep Neural Network (DNN) is nowadays widely used in many scenarios, including safety-critical applications such as autonomous driving. In this context, besides energy efficiency and performance, reliability plays a crucial role since a system failure can jeopardize human life. As with any other device, the reliability of hardware architectures running DNNs has to be evaluated, usually through costly fault injection campaigns. This paper explores the approximation and fault resiliency of DNN accelerators. We propose to use approximate (AxC) arithmetic circuits to agilely emulate errors in hardware without performing fault injection on the DNN. To allow fast evaluation of AxC DNN, we developed an efficient GPU-based simulation framework. Further, we propose a fine-grain analysis of fault resiliency by examining fault propagation and masking in networks

Mahdi Taheri, Saeideh Sheikhpour, Ali Mahani et al.

We introduce a novel logic style with self-checking capability to enhance hardware reliability at logic level. The proposed logic cells have two-rail inputs/outputs, and the functionality for each rail of outputs enables construction of faulttolerant configurable circuits. The AND and OR gates consist of 8 transistors based on CNFET technology, while the proposed XOR gate benefits from both CNFET and low-power MGDI technologies in its transistor arrangement. To demonstrate the feasibility of our new logic gates, we used an AES S-box implementation as the use case. The extensive simulation results using HSPICE indicate that the case-study circuit using on proposed gates has superior speed and power consumption compared to other implementations with error-detection capability

Mahdi Taheri, Mohammad Hasan Ahmadilivani, Maksim Jenihhin et al.

Nowadays, the extensive exploitation of Deep Neural Networks (DNNs) in safety-critical applications raises new reliability concerns. In practice, methods for fault injection by emulation in hardware are efficient and widely used to study the resilience of DNN architectures for mitigating reliability issues already at the early design stages. However, the state-of-the-art methods for fault injection by emulation incur a spectrum of time-, design- and control-complexity problems. To overcome these issues, a novel resiliency assessment method called APPRAISER is proposed that applies functional approximation for a non-conventional purpose and employs approximate computing errors for its interest. By adopting this concept in the resiliency assessment domain, APPRAISER provides thousands of times speed-up in the assessment process, while keeping high accuracy of the analysis. In this paper, APPRAISER is validated by comparing it with state-of-the-art approaches for fault injection by emulation in FPGA. By this, the feasibility of the idea is demonstrated, and a new perspective in resiliency evaluation for DNNs is opened.

Mohammad Hasan Ahmadilivani, Mahdi Taheri, Jaan Raik et al.

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.

Aneesh Balakrishnan, Thomas Lange, Maximilien Glorieux et al.

The paper is proposing a methodology for modeling a gate-level netlist using a Graph Convolutional Network (GCN). The model predicts the overall functional de-rating factors of sequential elements of a given circuit. In the preliminary phase of the work, the important goal is making a GCN which able to take a gate-level netlist as input information after transforming it into the Probabilistic Bayesian Graph in the form of Graph Modeling Language (GML). This part enables the GCN to learn the structural information of netlist in graph domains. In the second phase of the work, the modeled GCN trained with the a functional de-rating factor of a very low number of individual sequential elements (flip-flops). The third phase includes understanding of GCN models accuracy to model an arbitrary circuit netlist. The designed model was validated for two circuits. One is the IEEE 754 standard double precision floating point adder and the second one is the 10-Gigabit Ethernet MAC IEEE802.3 standard. The predicted results compared to the standard fault injection campaign results of the error called Single EventUpset (SEU). The validated results are graphically pictured in the form of the histogram and sorted probabilities and evaluated with the Confidence Interval (CI) metric between the predicted and simulated fault injection results.

Xinhui Lai, Maksim Jenihhin, Georgios Selimis et al.

Physical Unclonable Functions (PUFs) are gaining attention in the cryptography community because of the ability to efficiently harness the intrinsic variability in the manufacturing process. However, this means that they are noisy devices and require error correction mechanisms, e.g., by employing Fuzzy Extractors (FEs). Recent works demonstrated that applying FEs for error correction may enable new opportunities to break the PUFs if no countermeasures are taken. In this paper, we address an attack model on FEs hardware implementations and provide a solution for early identification of the timing Side-Channel Attack (SCA) vulnerabilities which can be exploited by physical fault injection. The significance of this work stems from the fact that FEs are an essential building block in the implementations of PUF-enabled devices. The information leaked through the timing side-channel during the error correction process can reveal the FE input data and thereby can endanger revealing secrets. Therefore, it is very important to identify the potential leakages early in the process during RTL design. Experimental results based on RTL analysis of several Bose-Chaudhuri-Hocquenghem (BCH) and Reed-Solomon decoders for PUF-enabled devices with FEs demonstrate the feasibility of the proposed methodology.

Xinhui Lai, Maksim Jenihhin, Jaan Raik et al.

A large number of crypto accelerators are being deployed with the widespread adoption of IoT. It is vitally important that these accelerators and other security hardware IPs are provably secure. Security is an extra functional requirement and hence many security verification tools are not mature. We propose an approach/flow-PASCAL-that works on RTL designs and discovers potential Timing Side-Channel Attack(SCA) vulnerabilities in them. Based on information flow analysis, this is able to identify Timing Disparate Security Paths that could lead to information leakage. This flow also (automatically) eliminates the information leakage caused by the timing channel. The insertion of a lightweight Compensator Block as balancing or compliance FSM removes the timing channel with minimum modifications to the design with no impact on the clock cycle time or combinational delay of the critical path in the circuit.

Ahmet Cagri Bagbaba, Maksim Jenihhin, Jaan Raik et al.

This work proposes a fault injection methodology where Hardware Description Language (HDL) code slicing is exploited to prune fault injection locations, thus enabling more efficient campaigns for safety mechanisms evaluation. In particular, the dynamic HDL slicing technique provides for a highly collapsed critical fault list and allows avoiding injections at redundant locations or time-steps. Experimental results show that the proposed methodology integrated into commercial tool flow doubles the simulation speed when comparing to the state-of-the-art industrial-grade EDA tool flows.

Maksim Jenihhin, Said Hamdioui, Matteo Sonza Reorda et al.

The recent trends for nanoelectronic computing systems include machine-to-machine communication in the era of Internet-of-Things (IoT) and autonomous systems, complex safety-critical applications, extreme miniaturization of implementation technologies and intensive interaction with the physical world. These set tough requirements on mutually dependent extra-functional design aspects. The H2020 MSCA ITN project RESCUE is focused on key challenges for reliability, security and quality, as well as related electronic design automation tools and methodologies. The objectives include both research advancements and cross-sectoral training of a new generation of interdisciplinary researchers. Notable interdisciplinary collaborative research results for the first half-period include novel approaches for test generation, soft-error and transient faults vulnerability analysis, cross-layer fault-tolerance and error-resilience, functional safety validation, reliability assessment and run-time management, HW security enhancement and initial implementation of these into holistic EDA tools.

Maksim Jenihhin, Matteo Sonza Reorda, Aneesh Balakrishnan et al.

The gigantic complexity and heterogeneity of today's advanced cyber-physical systems and systems of systems is multiplied by the use of avant-garde computing architectures to employ artificial intelligence based autonomy in the system. Here, the overall system's reliability comes along with requirements for fail-safe, fail-operational modes specific to the target applications of the autonomous system and adopted HW architectures. The paper makes an overview of reliability challenges for intelligence implementation in autonomous systems enabled by HW backbones such as neuromorphic architectures, approximate computing architectures, GPUs, tensor processing units (TPUs) and SoC FPGAs.