LATIDIA · Ciberseguridad

AutoML en línea: evaluación de ataques de envenenamiento en la estrategia de defensa de entrenamiento adversario en redes IoT

arXiv: 2610.05810v1Tipo de anuncio: nuevo Resumen: Los vectores de ataque de envenenamiento impulsados por aprendizaje automático (ML) son maniobras adversarias mediante las cuales un atacante inserta, corrompe o altera intencionalmente los datos de entrenamiento para distorsionar un

La noticia

arXiv:2610.05810v1 Announce Type: new Abstract: Machine learning (ML)-powered poisoning attack vectors are adversarial maneuvers whereby an attacker intentionally inserts, corrupts, or alters training data to distort an ML model's learning process. The objective is to diminish model efficacy, instill biases, induce misclassifications, or include concealed backdoors that may be attacked during implementation. In streaming contexts, poisoning attacks pose significant risks since models perpetually update based on incoming streams of data. An assailant may incrementally introduce harmful samples into this data stream, leading the model to assimilate erroneous features over time without timely identification. Therefore, this study is aimed at evaluating the efficacy of the adversarial training (AT) defense approach against