Data poisoning refers to a method of attack in which an adversary deliberately introduces misleading or harmful data into a aprendizado de máquina (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.
Em muitos sistemas de aprendizado de máquina systems, the quality and reliability of the dados de treinamento 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.
O envenenamento de dados pode assumir várias formas, incluindo:
- Troca de rótulos: Alterar os rótulos de certos pontos de dados para enganar o modelo durante o treinamento.
- Ataques de porta dos fundos: Inserting specific data patterns that cause the model to behave incorrectly only when those patterns are present.
- Adição de outliers: Introducing extreme or unusual data points that skew the model’s understanding of the normal distribuição de dados.
Mitigar o envenenamento de dados envolve múltiplas estratégias, como validação de dados robusta técnicas de validação de dados, techniques, detecção de anomalias, 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.