About This Blog

I’ll be sharing my thoughts on various topics including:

• Differential Privacy: Theory and practical applications
• Machine Learning: Privacy-preserving techniques in ML
• Research Insights: Learnings from academic and industry research

A Quick Demo

Here’s some code highlighting:

import numpy as np

def add_laplace_noise(value, sensitivity, epsilon):
    """Add Laplace noise for differential privacy."""
    scale = sensitivity / epsilon
    noise = np.random.laplace(0, scale)
    return value + noise

And some math for differential privacy:

Where \(\mathcal{M}\) is a randomized mechanism, and \(D, D'\) are neighboring databases.

Stay Tuned

More posts coming soon. Feel free to reach out if you have any questions or topics you’d like me to cover!