Data poisoning refers to a method of attack in which an adversary deliberately introduces misleading or harmful data into a aprendizaje automático (ML) training dataset. The objective of this malicious act is to compromise the integrity of the ML model, leading to incorrect predictions or classifications when the model is deployed in real-world applications.
En muchos sistemas de aprendizaje automático systems, the quality and reliability of the datos de entrenamiento are crucial for the model’s performance. When an attacker successfully implements data poisoning, they can manipulate the learning process by injecting biased or false information. This can result in a model that performs poorly, behaves unpredictably, or even serves the attacker’s goals by making specific predictions that benefit them.
El envenenamiento de datos puede adoptar varias formas, incluyendo:
- Cambio de etiquetas: Alterar las etiquetas de ciertos puntos de datos para engañar al modelo durante el entrenamiento.
- Ataques de puerta trasera: Inserting specific data patterns that cause the model to behave incorrectly only when those patterns are present.
- Adición de valores atípicos: Introducing extreme or unusual data points that skew the model’s understanding of the normal distribución de datos.
Mitigar el envenenamiento de datos implica múltiples estrategias, como técnicas robustas validación de datos techniques, detección de anomalías, and using diverse training datasets to minimize the impact of any single source of data corruption. Continuous monitoring and updating of models can also help to reduce the risk of data poisoning attacks.