Institution profile

Concordia University

Academic institutionnorthamerica · ca
Official website
Research library523linked papers
Opportunities0open roles
Selected work

Representative Papers

Real-Time Adaptive Anomaly Detection in Industrial IoT Environments

Dec 01, 2024IEEE Transactions on Network and Service Management

This work addresses the significant challenges posed by the dynamic and complex nature of high-dimensional heterogeneous data streams in Industrial Internet of Things (IIoT) environments for real-time anomaly detection. To this end, we propose a novel detection approach that integrates multi-source predictive modeling with an adaptive mechanism for concept drift. By continuously identifying distributional shifts and dynamically updating the underlying model, the method substantially enhances detection accuracy and robustness. Experimental evaluation on real-world IIoT datasets demonstrates that the proposed approach achieves an AUC of 89.71%, significantly outperforming current state-of-the-art methods while maintaining strong real-time performance, scalability, and computational efficiency.

11 citations1 influentialRead paper

Performance Optimization of Energy-Harvesting Underlay Cognitive Radio Networks Using Reinforcement Learning

Jun 19, 2023International Conference on Wireless Communications and Mobile Computing

In energy-constrained cognitive radio networks, secondary users (SUs) must coexist with primary users (PUs) while optimizing performance under stringent energy limitations. Method: This paper proposes a dynamic joint energy harvesting and data transmission decision framework. It innovatively treats PU communication signals—not as interference but as exploitable radio-frequency (RF) energy—and designs a dual-source adaptive energy harvesting mechanism. Integrating time-switching protocols with a deep Q-network (DQN), the framework jointly optimizes spectrum sensing, transceiver mode switching, and transmit power allocation. Contribution/Results: The proposed method significantly improves the SU’s average data rate, exhibits stable convergence, and consistently outperforms conventional benchmark strategies across diverse channel conditions and energy constraints. It establishes a novel paradigm for green, self-sustaining cognitive access by enabling SUs to autonomously harvest ambient RF energy from PU transmissions.

10 citationsRead paper

Neural Rank Collapse: Weight Decay and Small Within-Class Variability Yield Low-Rank Bias

Feb 06, 2024arXiv.org

This work investigates the origin of low-rank bias in deep neural networks and its connection to neural collapse. For general feedforward networks with nonlinear activations, we propose the “neural rank collapse” mechanism: weight decay jointly with intra-class variance in hidden layers drives rapid singular value decay across weight matrices, inducing progressive rank reduction. We establish, for the first time in nonlinear deep networks, a quantitative theoretical link between low-rank bias and neural collapse—extending beyond existing linear-network analyses. Our theory proves that the rank decay rate is proportional to the intra-class variance of the preceding layer’s hidden representations. Using singular value analysis, statistical modeling of latent-space distributions, and extensive experiments across architectures (ResNet, CNN), we empirically validate the mechanism. Furthermore, leveraging this insight, we achieve controllable rank compression of weight matrices by over 30% without sacrificing accuracy.

8 citationsRead paper

Towards Understanding Retrieval Accuracy and Prompt Quality in RAG Systems

Nov 29, 2024arXiv.org

The impact of key design decisions—RAG activation, retrieval granularity, and knowledge integration strategy—on RAG system performance remains poorly understood. Method: We conduct systematic ablation studies across three code/qa benchmarks and two state-of-the-art LLMs, quantitatively evaluating how document type, recall rate, document selection strategy, and prompt engineering jointly affect answer correctness and confidence via multi-dimensional analysis, cross-model/dataset comparison, and joint prompt-retrieval analysis. Contribution/Results: We identify precise interaction patterns and operational boundaries among these factors and propose nine actionable, empirically grounded guidelines for diagnosing and optimizing RAG failures. Our findings significantly improve RAG system stability, debuggability, and reliability, offering rigorous empirical evidence and a principled methodology to support the engineering deployment of LLM-augmented systems.

7 citationsRead paper

Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erd\H{o}s Problems

Jan 29, 2026

This work proposes a semi-autonomous discovery framework that integrates artificial intelligence with human expertise to investigate 700 mathematical conjectures labeled as “open” in Bloom’s Erdős Problem Database. Leveraging the Gemini large language model for natural language reasoning and automated literature comparison as an initial screening step, candidate solutions are subsequently evaluated by domain experts for correctness and novelty. The study reveals that many problems deemed “open” stem not from intrinsic difficulty but from challenges in literature retrieval—termed “information occlusion.” Among the 13 problems successfully resolved, five yielded novel AI-generated solutions, while eight were traced to previously published results. This research represents the first large-scale demonstration of human–AI collaborative verification in mathematical conjectures and highlights the risk of “unconscious plagiarism” inherent in AI-assisted scholarly discovery.

2 citationsRead paper
Recent publications

Latest Papers