Ouassim Karrakchou

Alaa Zniber, Ouassim Karrakchou, Mounir Ghogho

Early-exiting neural networks enable adaptive inference by allowing inputs to exit at intermediate classifiers, reducing computation for easy samples while maintaining high accuracy. In practice, exits can be trained sequentially by incrementally adding them to a shared backbone; however, this sequential training can cause newly introduced exits to interfere with previously learned ones, degrading the performance of earlier classifiers. We address this problem by retaining the knowledge embedded in existing exits while allowing new ones to specialize. We propose two alternative approaches that operate at different levels of the model. The first constrains learning by protecting parameters that are important for previously trained exits, while the second preserves the output distributions of earlier exits as the network adapts. These alternatives directly reflect the stability-plasticity trade-off studied in continual learning. Accordingly, we leverage \textit{Elastic Weight Consolidation} to constrain critical weights and \textit{Learning without Forgetting} to preserve output distributions. Experiments on standard benchmarks show that our approaches consistently improve early-exit performance, achieving higher accuracy over existing sequential training methods and significant performance speedups at low computational budgets.

Riccardo Miccini, Alaa Zniber, Clément Laroche et al.

Although deep learning has made strides in the field of deep noise suppression, leveraging deep architectures on resource-constrained devices still proved challenging. Therefore, we present an early-exiting model based on nsNet2 that provides several levels of accuracy and resource savings by halting computations at different stages. Moreover, we adapt the original architecture by splitting the information flow to take into account the injected dynamism. We show the trade-offs between performance and computational complexity based on established metrics.

Alaa Zniber, Ouassim Karrakchou, Mounir Ghogho

Internet of Things (IoT) sensors are nowadays heavily utilized in various real-world applications ranging from wearables to smart buildings passing by agrotechnology and health monitoring. With the huge amounts of data generated by these tiny devices, Deep Learning (DL) models have been extensively used to enhance them with intelligent processing. However, with the urge for smaller and more accurate devices, DL models became too heavy to deploy. It is thus necessary to incorporate the hardware's limited resources in the design process. Therefore, inspired by the human brain known for its efficiency and low power consumption, we propose a shallow bidirectional network based on predictive coding theory and dynamic early exiting for halting further computations when a performance threshold is surpassed. We achieve comparable accuracy to VGG-16 in image classification on CIFAR-10 with fewer parameters and less computational complexity.

Ouassim Karrakchou, Alaa Zniber, Anass Sebbar et al.

Distributed Denial of Service (DDoS) attacks pose a persistent threat to network security, requiring timely and scalable mitigation strategies. In this paper, we propose a novel collaborative architecture that integrates a P4-programmable data plane with an SDN control plane to enable real-time DDoS detection and response. At the core of our approach is a split early-exit neural network that performs partial inference in the data plane using a quantized Convolutional Neural Network (CNN), while deferring uncertain cases to a Gated Recurrent Unit (GRU) module in the control plane. This design enables high-speed classification at line rate with the ability to escalate more complex flows for deeper analysis. Experimental evaluation using real-world DDoS datasets demonstrates that our approach achieves high detection accuracy with significantly reduced inference latency and control plane overhead. These results highlight the potential of tightly coupled ML-P4-SDN systems for efficient, adaptive, and low-latency DDoS defense.

Alaa Zniber, Arne Symons, Ouassim Karrakchou et al.

Advancements in high-performance computing and cloud technologies have enabled the development of increasingly sophisticated Deep Learning (DL) models. However, the growing demand for embedded intelligence at the edge imposes stringent computational and energy constraints, challenging the deployment of these large-scale models. Early Exiting Neural Networks (EENN) have emerged as a promising solution, allowing dynamic termination of inference based on input complexity to enhance efficiency. Despite their potential, EENN performance is highly influenced by the heterogeneity of edge accelerators and the constraints imposed by quantization, affecting accuracy, energy efficiency, and latency. Yet, research on the automatic optimization of EENN design for edge hardware remains limited. To bridge this gap, we propose a hardware-aware Neural Architecture Search (NAS) framework that systematically integrates the effects of quantization and hardware resource allocation to optimize the placement of early exit points within a network backbone. Experimental results on the CIFAR-10 dataset demonstrate that our NAS framework can discover architectures that achieve over a 50\% reduction in computational costs compared to conventional static networks, making them more suitable for deployment in resource-constrained edge environments.

Meryem Amaouche, Ouassim Karrakchou, Mounir Ghogho et al.

Bladder cancer ranks within the top 10 most diagnosed cancers worldwide and is among the most expensive cancers to treat due to the high recurrence rates which require lifetime follow-ups. The primary tool for diagnosis is cystoscopy, which heavily relies on doctors' expertise and interpretation. Therefore, annually, numerous cases are either undiagnosed or misdiagnosed and treated as urinary infections. To address this, we suggest a deep learning approach for bladder cancer detection and segmentation which combines CNNs with a lightweight positional-encoding-free transformer and dual attention gates that fuse self and spatial attention for feature enhancement. The architecture suggested in this paper is efficient making it suitable for medical scenarios that require real time inference. Experiments have proven that this model addresses the critical need for a balance between computational efficiency and diagnostic accuracy in cystoscopic imaging as despite its small size it rivals large models in performance.

Saad Mokssit, Ouassim Karrakchou, Alejandro Mousist et al.

Early-exit neural networks reduce inference cost by enabling confident predictions at intermediate layers. However, joint training often leads to gradient interference, with deeper classifiers dominating optimization. We propose Confidence-Gated Training (CGT), a paradigm that conditionally propagates gradients from deeper exits only when preceding exits fail. This encourages shallow classifiers to act as primary decision points while reserving deeper layers for harder inputs. By aligning training with the inference-time policy, CGT mitigates overthinking, improves early-exit accuracy, and preserves efficiency. Experiments on the Indian Pines and Fashion-MNIST benchmarks show that CGT lowers average inference cost while improving overall accuracy, offering a practical solution for deploying deep models in resource-constrained environments.

Nouhaila Fraihi, Ouassim Karrakchou, Mounir Ghogho

Accurate classification of respiratory sounds requires deep learning models that effectively capture fine-grained acoustic features and long-range temporal dependencies. Convolutional Neural Networks (CNNs) are well-suited for extracting local time-frequency patterns but are limited in modeling global context. In contrast, transformer-based models can capture long-range dependencies, albeit with higher computational demands. To address these limitations, we propose a compact CNN-Temporal Self-Attention (CNN-TSA) network that integrates lightweight self-attention into an efficient CNN backbone. Central to our approach is a Frequency Band Selection (FBS) module that suppresses noisy and non-informative frequency regions, substantially improving accuracy and reducing FLOPs by up to 50%. We also introduce age-specific models to enhance robustness across diverse patient groups. Evaluated on the SPRSound-2022/2023 and ICBHI-2017 lung sound datasets, CNN-TSA with FBS sets new benchmarks on SPRSound and achieves state-of-the-art performance on ICBHI, all with a significantly smaller computational footprint. Furthermore, integrating FBS into an existing transformer baseline yields a new record on ICBHI, confirming FBS as an effective drop-in enhancement. These results demonstrate that our framework enables reliable, real-time respiratory sound analysis suitable for deployment in resource-constrained settings.