MINT: Deep Network Compression via Mutual Information-based Neuron Trimming

Ganesh, Madan Ravi; Corso, Jason J.; Sekeh, Salimeh Yasaei

doi:10.1109/icpr48806.2021.9412590

Cited by 5 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, deep learning approaches, such as autoencoders [16], have demonstrated their capacity to uncover intricate patterns within large-scale power system data, aiding in feature selection for DSA. At the same time, methods like transfer learning [17], attribute selection techniques [18], and mutual information-based approaches [19,20] have been applied to identify the most influential features. These techniques help reduce computational complexity, improve model interpretability, and enhance the performance of data-driven models.…”

Section: Introductionmentioning

confidence: 99%

Data‐driven power system dynamic security assessment under adversarial attacks: Risk warning based interpretation analysis and mitigation

Chen,

Ren,

et al. 2023

IET Energy Syst Integration

View full text Add to dashboard Cite

Power system dynamic security assessment (DSA) has long been essential for protecting the system from the risk of cascading failures and wide‐spread blackouts. The machine learning (ML) based data‐driven strategy is promising due to its real‐time computation speed and knowledge discovery capacity. However, ML algorithms are found to be vulnerable against well‐designed malicious input samples that can lead to wrong outputs. Adversarial attacks are implemented to measure the vulnerability of the trained ML models. Specifically, the targets of attacks are identified by interpretation analysis that the data features with large SHAP values will be assigned with perturbations. The proposed method has the superiority that an instance‐based DSA method is established with interpretation of the ML models, where effective adversarial attacks and its mitigation countermeasure are developed by assigning the perturbations on features with high importance. Later, these generated adversarial examples are employed for adversarial training and mitigation. The simulation results present that the model accuracy and robustness vary with the quantity of adversarial examples used, and there is not necessarily a trade‐off between these two indicators. Furthermore, the rate of successful attacks increases when a greater bound of perturbations is permitted. By this method, the correlation between model accuracy and robustness can be clearly stated, which will provide considerable assistance in decision making.

show abstract