What are the key techniques used in adversarial machine learning as discussed in the chapter?

The key techniques in adversarial machine learning discussed include evasion attacks, poisoning attacks, and model extraction attacks.

Adversarial machine learning techniques are sophisticated methods used to manipulate AI models by exploiting their vulnerabilities. Evasion attacks occur during the inference phase, where attackers subtly alter input data to deceive the model into making incorrect classifications. These are often used to bypass fraud detection systems by making small, strategic changes to transaction patterns. Poisoning attacks target the training phase by injecting malicious data points into the training dataset, which can bias future model predictions. Model extraction attacks involve stealing proprietary algorithms by querying API endpoints, allowing attackers to replicate a model without access to its training data. These techniques pose significant risks to financial institutions as they can be automated and scaled across numerous accounts or transactions.