Adversarial attack

{
  "id": "MLC-T-OAG-000065",
  "term": "Adversarial attack",
  "field": "Oil and gas",
  "definition": "A malicious input designed to deceive a machine learning model, causing it to misclassify data or make incorrect predictions. These attacks often involve subtle perturbations to input data that are imperceptible to humans but significantly alter the model's output. In oil and gas, such attacks could target predictive maintenance systems or seismic interpretation models, leading to operational errors.",
  "subject": "Digital",
  "url": "https://mlchart.com/terminology/oil-and-gas/adversarial-attack/"
}