arXiv subjects

Saeed Bagheri Shouraki

Publications and source records attributed to Saeed Bagheri Shouraki.

At least 19 recordsLinked to original sources

arXiv2026

Ensemble-Guided Distillation for Compact and Robust Acoustic Scene Classification on Edge Devices

We present a compact, quantization-ready acoustic scene classification (ASC) framework that couples an efficient student network with a learned teacher ensemble and knowledge distillation. The student backbone uses stacked depthwise-separable "expand-depthwise-project" blocks with global response normalization to stabilize training and improve robustness to device and noise variability, while a global pooling head yields class logits for efficient edge inference. To inject richer inductive bias, we assemble a diverse set of teacher models and learn two complementary fusion heads: z1, which predicts per-teacher mixture weights using a student-style backbone, and z2, a lightweight MLP that performs per-class logit fusion. The student is distilled from the ensemble via temperature-scaled soft targets combined with hard labels, enabling it to approximate the ensemble's decision geometry with a single compact model. Evaluated on the TAU Urban Acoustic Scenes 2022 Mobile benchmark, our approach achieves state-of-the-art (SOTA) results on the TAU dataset under matched edge-deployment constraints, demonstrating strong performance and practicality for mobile ASC.

cs.SD↗

arXiv2026

Parallel Data Processing in Quantum Machine Learning

We propose a Quantum Machine Learning (QML) framework that applies the core design principle of quantum algorithms-superposition, oracle, and interference-to accelerate training. Building on the structural analogy between feature extraction in foundational quantum algorithms and parameter optimization in QML, we reformulate the training process to leverage quantum parallelism: all training samples are encoded into a quantum superposition, processed through a parameterized quantum circuit, and classified via an interferometer module that implements quantum interference across the dataset. This architectural reformulation reduces the theoretical complexity of loss function evaluation from $O(N^{2})$ in conventional QML training to $O(N)$, where $N$ is the dataset size. Numerical simulations on multiple binary and multi-class classification datasets (with up to $N=128$ samples) demonstrate that our method achieves classification accuracies comparable to conventional circuits while reducing the number of quantum circuit executions per cost function evaluation from $N$ to 1. This represents a near $N$-fold reduction in quantum overhead per training iteration, reducing the required circuit executions without loss of accuracy. These results highlight the potential of quantum algorithmic design principles as a scalable pathway to efficient QML implementations.

quant-ph↗

arXiv2026

CLIP-Joint-Detect: End-to-End Joint Training of Object Detectors with Contrastive Vision-Language Supervision

Conventional object detectors rely on cross-entropy classification, which can be vulnerable to class imbalance and label noise. We propose CLIP-Joint-Detect, a simple and detector-agnostic framework that integrates CLIP-style contrastive vision-language supervision through end-to-end joint training. A lightweight parallel head projects region or grid features into the CLIP embedding space and aligns them with learnable class-specific text embeddings via InfoNCE contrastive loss and an auxiliary cross-entropy term, while all standard detection losses are optimized simultaneously. The approach applies seamlessly to both two-stage and one-stage architectures. We validate it on Pascal VOC 2007+2012 using Faster R-CNN and on the large-scale MS COCO 2017 benchmark using modern YOLO detectors (YOLOv11), achieving consistent and substantial improvements while preserving real-time inference speed. Extensive experiments and ablations demonstrate that joint optimization with learnable text embeddings markedly enhances closed-set detection performance across diverse architectures and datasets.

cs.CV↗

arXiv2025

GradID: Adversarial Detection via Intrinsic Dimensionality of Gradients

Despite their remarkable performance, deep neural networks exhibit a critical vulnerability: small, often imperceptible, adversarial perturbations can lead to drastically altered model predictions. Given the stringent reliability demands of applications such as medical diagnosis and autonomous driving, robust detection of such adversarial attacks is paramount. In this paper, we investigate the geometric properties of a model's input loss landscape. We analyze the Intrinsic Dimensionality (ID) of the model's gradient parameters, which quantifies the minimal number of coordinates required to describe the data points on their underlying manifold. We reveal a distinct and consistent difference in the ID for natural and adversarial data, which forms the basis of our proposed detection method. We validate our approach across two distinct operational scenarios. First, in a batch-wise context for identifying malicious data groups, our method demonstrates high efficacy on datasets like MNIST and SVHN. Second, in the critical individual-sample setting, we establish new state-of-the-art results on challenging benchmarks such as CIFAR-10 and MS COCO. Our detector significantly surpasses existing methods against a wide array of attacks, including CW and AutoAttack, achieving detection rates consistently above 92\% on CIFAR-10. The results underscore the robustness of our geometric approach, highlighting that intrinsic dimensionality is a powerful fingerprint for adversarial detection across diverse datasets and attack strategies.

cs.LG↗

arXiv2025

Human-Centric Anomaly Detection in Surveillance Videos Using YOLO-World and Spatio-Temporal Deep Learning

Anomaly detection in surveillance videos remains a challenging task due to the diversity of abnormal events, class imbalance, and scene-dependent visual clutter. To address these issues, we propose a robust deep learning framework that integrates human-centric preprocessing with spatio-temporal modeling for multi-class anomaly classification. Our pipeline begins by applying YOLO-World - an open-vocabulary vision-language detector - to identify human instances in raw video clips, followed by ByteTrack for consistent identity-aware tracking. Background regions outside detected bounding boxes are suppressed via Gaussian blurring, effectively reducing scene-specific distractions and focusing the model on behaviorally relevant foreground content. The refined frames are then processed by an ImageNet-pretrained InceptionV3 network for spatial feature extraction, and temporal dynamics are captured using a bidirectional LSTM (BiLSTM) for sequence-level classification. Evaluated on a five-class subset of the UCF-Crime dataset (Normal, Burglary, Fighting, Arson, Explosion), our method achieves a mean test accuracy of 92.41% across three independent trials, with per-class F1-scores consistently exceeding 0.85. Comprehensive evaluation metrics - including confusion matrices, ROC curves, and macro/weighted averages - demonstrate strong generalization and resilience to class imbalance. The results confirm that foreground-focused preprocessing significantly enhances anomaly discrimination in real-world surveillance scenarios.

cs.CV↗

arXiv2025

Secure Diagnostics: Adversarial Robustness Meets Clinical Interpretability

Deep neural networks for medical image classification often fail to generalize consistently in clinical practice due to violations of the i.i.d. assumption and opaque decision-making. This paper examines interpretability in deep neural networks fine-tuned for fracture detection by evaluating model performance against adversarial attack and comparing interpretability methods to fracture regions annotated by an orthopedic surgeon. Our findings prove that robust models yield explanations more aligned with clinically meaningful areas, indicating that robustness encourages anatomically relevant feature prioritization. We emphasize the value of interpretability for facilitating human-AI collaboration, in which models serve as assistants under a human-in-the-loop paradigm: clinically plausible explanations foster trust, enable error correction, and discourage reliance on AI for high-stakes decisions. This paper investigates robustness and interpretability as complementary benchmarks for bridging the gap between benchmark performance and safe, actionable clinical deployment.

cs.CV↗

arXiv2024

DFT-Based Adversarial Attack Detection in MRI Brain Imaging: Enhancing Diagnostic Accuracy in Alzheimer's Case Studies

Recent advancements in deep learning, particularly in medical imaging, have significantly propelled the progress of healthcare systems. However, examining the robustness of medical images against adversarial attacks is crucial due to their real-world applications and profound impact on individuals' health. These attacks can result in misclassifications in disease diagnosis, potentially leading to severe consequences. Numerous studies have explored both the implementation of adversarial attacks on medical images and the development of defense mechanisms against these threats, highlighting the vulnerabilities of deep neural networks to such adversarial activities. In this study, we investigate adversarial attacks on images associated with Alzheimer's disease and propose a defensive method to counteract these attacks. Specifically, we examine adversarial attacks that employ frequency domain transformations on Alzheimer's disease images, along with other well-known adversarial attacks. Our approach utilizes a convolutional neural network (CNN)-based autoencoder architecture in conjunction with the two-dimensional Fourier transform of images for detection purposes. The simulation results demonstrate that our detection and defense mechanism effectively mitigates several adversarial attacks, thereby enhancing the robustness of deep neural networks against such vulnerabilities.

eess.IV↗

arXiv2023

Face: Fast, Accurate and Context-Aware Audio Annotation and Classification

This paper presents a context-aware framework for feature selection and classification procedures to realize a fast and accurate audio event annotation and classification. The context-aware design starts with exploring feature extraction techniques to find an appropriate combination to select a set resulting in remarkable classification accuracy with minimal computational effort. The exploration for feature selection also embraces an investigation of audio Tempo representation, an advantageous feature extraction method missed by previous works in the environmental audio classification research scope. The proposed annotation method considers outlier, inlier, and hard-to-predict data samples to realize context-aware Active Learning, leading to the average accuracy of 90% when only 15% of data possess initial annotation. Our proposed algorithm for sound classification obtained average prediction accuracy of 98.05% on the UrbanSound8K dataset. The notebooks containing our source codes and implementation results are available at https://github.com/gitmehrdad/FACE.

cs.SD↗

arXiv2022

Encrypted Internet traffic classification using a supervised Spiking Neural Network

Internet traffic recognition is an essential tool for access providers since recognizing traffic categories related to different data packets transmitted on a network help them define adapted priorities. That means, for instance, high priority requirements for an audio conference and low ones for a file transfer, to enhance user experience. As internet traffic becomes increasingly encrypted, the mainstream classic traffic recognition technique, payload inspection, is rendered ineffective. This paper uses machine learning techniques for encrypted traffic classification, looking only at packet size and time of arrival. Spiking neural networks (SNN), largely inspired by how biological neurons operate, were used for two reasons. Firstly, they are able to recognize time-related data packet features. Secondly, they can be implemented efficiently on neuromorphic hardware with a low energy footprint. Here we used a very simple feedforward SNN, with only one fully-connected hidden layer, and trained in a supervised manner using the newly introduced method known as Surrogate Gradient Learning. Surprisingly, such a simple SNN reached an accuracy of 95.9% on ISCX datasets, outperforming previous approaches. Besides better accuracy, there is also a very significant improvement on simplicity: input size, number of neurons, trainable parameters are all reduced by one to four orders of magnitude. Next, we analyzed the reasons for this good accuracy. It turns out that, beyond spatial (i.e. packet size) features, the SNN also exploits temporal ones, mostly the nearly synchronous (within a 200ms range) arrival times of packets with certain sizes. Taken together, these results show that SNNs are an excellent fit for encrypted internet traffic classification: they can be more accurate than conventional artificial neural networks (ANN), and they could be implemented efficiently on low power embedded systems.

cs.LG↗

arXiv2019

Optimization of dynamic mobile robot path planning based on evolutionary methods

This paper presents evolutionary methods for optimization in dynamic mobile robot path planning. In dynamic mobile path planning, the goal is to find an optimal feasible path from starting point to target point with various obstacles, as well as smoothness and safety in the proposed path. Pattern search (PS) algorithm, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are used to find an optimal path for mobile robots to reach to target point with obstacle avoidance. For showing the success of the proposed method, first they are applied to two different paths with a dynamic environment in obstacles. The first results show that the PSO algorithms are converged and minimize the objective function better that the others, while PS has the lower time compared to other algorithms in the initial and modified environment. The second test path is in the z-type environment that we compare the mentioned algorithms on it. Also in this environment, the same result is repeated.

cs.RO↗

arXiv2017

A Fusion-based Gender Recognition Method Using Facial Images

This paper proposes a fusion-based gender recognition method which uses facial images as input. Firstly, this paper utilizes pre-processing and a landmark detection method in order to find the important landmarks of faces. Thereafter, four different frameworks are proposed which are inspired by state-of-the-art gender recognition systems. The first framework extracts features using Local Binary Pattern (LBP) and Principal Component Analysis (PCA) and uses back propagation neural network. The second framework uses Gabor filters, PCA, and kernel Support Vector Machine (SVM). The third framework uses lower part of faces as input and classifies them using kernel SVM. The fourth framework uses Linear Discriminant Analysis (LDA) in order to classify the side outline landmarks of faces. Finally, the four decisions of frameworks are fused using weighted voting. This paper takes advantage of both texture and geometrical information, the two dominant types of information in facial gender recognition. Experimental results show the power and effectiveness of the proposed method. This method obtains recognition rate of 94% for neutral faces of FEI face dataset, which is equal to state-of-the-art rate for this dataset.

cs.CV↗

arXiv2016

Fast IDS Computing System Method and its Memristor Crossbar-based Hardware Implementation

Active Learning Method (ALM) is one of the powerful tools in soft computing that is inspired by human brain capabilities in processing complicated information. ALM, which is in essence an adaptive fuzzy learning method, models a Multi-Input Single-Output (MISO) system with several Single-Input Single-Output (SISO) subsystems. Ink Drop Spread (IDS) operator, which is the main processing engine of this method, extracts useful features from the data without complicated computations and provides stability and convergence as well. Despite great performance of ALM in applications such as classification, clustering, and modelling, an efficient hardware implementation has remained a challenging problem. Large amount of memory required to store the information of IDS planes as well as the high computational cost of the IDS computing system are two main barriers to ALM becoming more popular. In this paper, a novel learning method is proposed based on the idea of IDS, but with a novel approach that eliminates the computational cost of IDS operator. Unlike traditional approaches, our proposed method finds functions to describe the IDS plane that eliminates the need for large amount of memory to a great extent. Narrow Path and Spread, which are two main features used in the inference engine of ALM, are then extracted from IDS planes with minimum amount of memory usage and power consumption. Our proposed algorithm is fully compatible with memristor-crossbar implementation that leads to a significant decrease in the number of required memristors (from O(n^2) to O(3n)). Simpler algorithm and higher speed make our algorithm suitable for applications where real-time process, low-cost and small implementation are paramount. Applications in clustering and function approximation are provided, which reveals the effective performance of our proposed algorithm.

cs.ET↗

arXiv2013

Using memristor crossbar structure to implement a novel adaptive real time fuzzy modeling algorithm

Although fuzzy techniques promise fast meanwhile accurate modeling and control abilities for complicated systems, different difficulties have been re-vealed in real situation implementations. Usually there is no escape of it-erative optimization based on crisp domain algorithms. Recently memristor structures appeared promising to implement neural network structures and fuzzy algorithms. In this paper a novel adaptive real-time fuzzy modeling algorithm is proposed which uses active learning method concept to mimic recent understandings of right brain processing techniques. The developed method is based on processing fuzzy numbers to provide the ability of being sensitive to each training data point to expand the knowledge tree leading to plasticity while used defuzzification technique guaranties enough stability. An outstanding characteristic of the proposed algorithm is its consistency to memristor crossbar hardware processing concepts. An analog implemen-tation of the proposed algorithm on memristor crossbars structure is also introduced in this paper. The effectiveness of the proposed algorithm in modeling and pattern recognition tasks is verified by means of computer simulations

cs.AI↗

arXiv2013

Generating Motion Patterns Using Evolutionary Computation in Digital Soccer

Dribbling an opponent player in digital soccer environment is an important practical problem in motion planning. It has special complexities which can be generalized to most important problems in other similar Multi Agent Systems. In this paper, we propose a hybrid computational geometry and evolutionary computation approach for generating motion trajectories to avoid a mobile obstacle. In this case an opponent agent is not only an obstacle but also one who tries to harden dribbling procedure. One characteristic of this approach is reducing process cost of online stage by transferring it to offline stage which causes increment in agents' performance. This approach breaks the problem into two offline and online stages. During offline stage the goal is to find desired trajectory using evolutionary computation and saving it as a trajectory plan. A trajectory plan consists of nodes which approximate information of each trajectory plan. In online stage, a linear interpolation along with Delaunay triangulation in xy-plan is applied to trajectory plan to retrieve desired action.

cs.AI↗

arXiv2012

Neuro-Fuzzy Computing System with the Capacity of Implementation on Memristor-Crossbar and Optimization-Free Hardware Training

In this paper, first we present a new explanation for the relation between logical circuits and artificial neural networks, logical circuits and fuzzy logic, and artificial neural networks and fuzzy inference systems. Then, based on these results, we propose a new neuro-fuzzy computing system which can effectively be implemented on the memristor-crossbar structure. One important feature of the proposed system is that its hardware can directly be trained using the Hebbian learning rule and without the need to any optimization. The system also has a very good capability to deal with huge number of input-out training data without facing problems like overtraining.

cs.NE↗

arXiv2012

Bottleneck of using single memristor as a synapse and its solution

It is now widely accepted that memristive devices are perfect candidates for the emulation of biological synapses in neuromorphic systems. This is mainly because of the fact that like the strength of synapse, memristance of the memristive device can be tuned actively (e.g., by the application of volt- age or current). In addition, it is also possible to fabricate very high density of memristive devices (comparable to the number of synapses in real biological system) through the nano-crossbar structures. However, in this paper we will show that there are some problems associated with memristive synapses (memristive devices which are playing the role of biological synapses). For example, we show that the variation rate of the memristance of memristive device depends completely on the current memristance of the device and therefore it can change significantly with time during the learning phase. This phenomenon can degrade the performance of learning methods like Spike Timing-Dependent Plasticity (STDP) and cause the corresponding neuromorphic systems to become unstable. Finally, at the end of this paper, we illustrate that using two serially connected memristive devices with different polarities as a synapse can somewhat fix the aforementioned problem.

cs.NE↗

arXiv2011

Memristive fuzzy edge detector

Fuzzy inference systems always suffer from the lack of efficient structures or platforms for their hardware implementation. In this paper, we tried to overcome this problem by proposing new method for the implementation of those fuzzy inference systems which use fuzzy rule base to make inference. To achieve this goal, we have designed a multi-layer neuro-fuzzy computing system based on the memristor crossbar structure by introducing some new concepts like fuzzy minterms. Although many applications can be realized through the use of our proposed system, in this study we show how the fuzzy XOR function can be constructed and how it can be used to extract edges from grayscale images. Our memristive fuzzy edge detector (implemented in analog form) compared with other common edge detectors has this advantage that it can extract edges of any given image all at once in real-time.

cs.NE↗

arXiv2011

New S-norm and T-norm Operators for Active Learning Method

Active Learning Method (ALM) is a soft computing method used for modeling and control based on fuzzy logic. All operators defined for fuzzy sets must serve as either fuzzy S-norm or fuzzy T-norm. Despite being a powerful modeling method, ALM does not possess operators which serve as S-norms and T-norms which deprive it of a profound analytical expression/form. This paper introduces two new operators based on morphology which satisfy the following conditions: First, they serve as fuzzy S-norm and T-norm. Second, they satisfy Demorgans law, so they complement each other perfectly. These operators are investigated via three viewpoints: Mathematics, Geometry and fuzzy logic.

cs.AI↗