arXiv subjects

Ahmed Shafee

Publications and source records attributed to Ahmed Shafee.

5 recordsLinked to original sources

arXiv2026

When Proxy Prediction Becomes Equation Reconstruction: Diagnostics and Residual Learning for Factor-Derived Proxy Supervision

Scientific machine learning often relies on proxy targets computed from known domain factors when direct observations are limited. When those same factors are used as model inputs, however, high predictive accuracy may reflect reconstruction of the proxy-generating equation rather than robustness to degraded factor information. We study this problem in RUSLE-derived soil-loss proxy prediction under controlled degradation of the soil-erodibility factor $K$. We introduce a diagnostic framework that combines degraded-formula references, classical tree-based baselines, matched direct and formula-feature predictors, contextual ablations, tail-error analysis, and degradation robustness scoring. We then propose RASPL, a formula-preserving residual framework that retains the degraded formula estimate as the prediction anchor and learns an adaptively gated contextual correction. RASPL substantially outperforms matched direct prediction and provides stronger degradation and tail robustness than treating the formula estimate as an ordinary input feature. Within RASPL, a compact statistical encoder achieves the highest macro-averaged $R^2$ and lowest computational cost, whereas a convolutional encoder achieves the strongest degradation robustness and lowest Tail95 mean absolute error (MAE). These results establish formula preservation as the central design principle for robust learning from factor-derived proxy targets.

cs.LG↗

arXiv2025

APThreatHunter: An automated planning-based threat hunting framework

Cyber attacks threaten economic interests, critical infrastructure, and public health and safety. To counter this, entities adopt cyber threat hunting, a proactive approach that involves formulating hypotheses and searching for attack patterns within organisational networks. Automating cyber threat hunting presents challenges, particularly in generating hypotheses, as it is a manually created and confirmed process, making it time-consuming. To address these challenges, we introduce APThreatHunter, an automated threat hunting solution that generates hypotheses with minimal human intervention, eliminating analyst bias and reducing time and cost. This is done by presenting possible risks based on the system's current state and a set of indicators to indicate whether any of the detected risks are happening or not. We evaluated APThreatHunter using real-world Android malware samples, and the results revealed the practicality of using automated planning for goal hypothesis generation in cyber threat hunting activities.

cs.CR↗

arXiv2020

Detection of Lying Electrical Vehicles in Charging Coordination Application Using Deep Learning

The simultaneous charging of many electric vehicles (EVs) stresses the distribution system and may cause grid instability in severe cases. The best way to avoid this problem is by charging coordination. The idea is that the EVs should report data (such as state-of-charge (SoC) of the battery) to run a mechanism to prioritize the charging requests and select the EVs that should charge during this time slot and defer other requests to future time slots. However, EVs may lie and send false data to receive high charging priority illegally. In this paper, we first study this attack to evaluate the gains of the lying EVs and how their behavior impacts the honest EVs and the performance of charging coordination mechanism. Our evaluations indicate that lying EVs have a greater chance to get charged comparing to honest EVs and they degrade the performance of the charging coordination mechanism. Then, an anomaly based detector that is using deep neural networks (DNN) is devised to identify the lying EVs. To do that, we first create an honest dataset for charging coordination application using real driving traces and information revealed by EV manufacturers, and then we also propose a number of attacks to create malicious data. We trained and evaluated two models, which are the multi-layer perceptron (MLP) and the gated recurrent unit (GRU) using this dataset and the GRU detector gives better results. Our evaluations indicate that our detector can detect lying EVs with high accuracy and low false positive rate.

cs.CR↗

arXiv2020

Mimic Learning to Generate a Shareable Network Intrusion Detection Model

Purveyors of malicious network attacks continue to increase the complexity and the sophistication of their techniques, and their ability to evade detection continues to improve as well. Hence, intrusion detection systems must also evolve to meet these increasingly challenging threats. Machine learning is often used to support this needed improvement. However, training a good prediction model can require a large set of labelled training data. Such datasets are difficult to obtain because privacy concerns prevent the majority of intrusion detection agencies from sharing their sensitive data. In this paper, we propose the use of mimic learning to enable the transfer of intrusion detection knowledge through a teacher model trained on private data to a student model. This student model provides a mean of publicly sharing knowledge extracted from private data without sharing the data itself. Our results confirm that the proposed scheme can produce a student intrusion detection model that mimics the teacher model without requiring access to the original dataset.

cs.CR↗

arXiv2003

Investigation by physical methods of the possible role of telomeres in DNA in aging process

The interaction energies between the different types of bases of a single strand of DNA molecule have been calculated. Using these original values of energies the harmonic behavior of a number of base patterns of DNA has been studied. In view of the great interest aroused by the discovery of the role of the telomere segment of the DNA in the replication process and its possible link with the aging process, we have investigated, with simple models, the harmonic behavior of the telomeric pattern of bases as well as the thermodynamic response in the biological system. With these results a conclusion on the probable role of the telomeric pattern on aging has also been drawn. Here the calculated values of harmonic frequencies of the telomeric pattern of bases and of other possible patterns show that the telomeric pattern is associated with the highest vibrational frequency among all patterns of base combinations at the replication end of DNA. This seems to suggest that due to the existing telomeric pattern being closest to the frequencies of the electromagnetic radiation coming from sunlight, resonance of the telomeric frequency with such radiation may be responsible for damage to the reproductive ability of the cells and consequent aging and other problems. On the other hand in the last part of this work we have calculated the thermal vibrational amplitudes of the telomeric pattern and other possible patterns which show that the amplitude for the telomeric pattern is the least, and this suggests that the telomeric pattern is more mechanically and thermally stable than other possible patterns in the biological environment against damage from ordinary heat and mechanical effects.

physics.bio-ph↗