One of the most important subfields in Machine Learning is Privacy-Preserving ML. If you are interested in AI Safety, you should pay attention to it. Today we are going to talk about Differential Privacy.
Differential privacy (DP) provides a quantifiable privacy guarantee by ensuring that no person’s data significantly affects the probability of any outcome. W/o DP adversarial actors might be able to reconstruct training data samples (your personal information) by analyzing the model. Yikes!!!
Fortunately, the authors of the paper, “Privacy Auditing with One (1) Training Run”, present one of the best ways to quantify your pipeline privacy. Their work, “auditing scheme requires minimal assumptions about the algorithm and can be applied in the black-box or white-box setting.”
We propose a scheme for auditing differentially private machine learning systems with a single training run. This exploits the parallelism of being able to add or remove multiple training examples independently. We analyze this using the connection between differential privacy and statistical generalization, which avoids the cost of group privacy. Our auditing scheme requires minimal assumptions about the algorithm and can be applied in the black-box or white-box setting.
-The contriutions
Their work reminds me of the algorithm for permutation-based feature importance. To quote the authors, “We identify m data points (i.e., training examples or “canaries”) to either include or exclude and we flip m independent unbiased coins to decide which of them to include or exclude. We then run the algorithm on the randomly selected dataset. Based on the output of the algorithm, the auditor “guesses” whether or not each data point was included or excluded (or it can abstain from guessing for some data points). We obtain a lower bound on the privacy parameters from the fraction of guesses that were correct.”
If you are an ML Engineer, I highly recommend looking into their publication over here: https://arxiv.org/abs/2305.08846
For more details, sign up for my free AI Newsletter, AI Made Simple. AI Made Simple- https://artificialintelligencemadesimple.substack.com/
If you liked this article and wish to share it, please refer to the following guidelines.
If you find my writing useful and would like to support my writing- please consider becoming a premium member of my cult by subscribing below. Subscribing gives you access to a lot more content and enables me to continue writing. This will cost you 400 INR (5 USD) monthly or 4000 INR (50 USD) per year and comes with a 60-day, complete refund policy. Understand the newest developments and develop your understanding of the most important ideas, all for the price of a cup of coffee.
Reach out to me
Use the links below to check out my other content, learn more about tutoring, reach out to me about projects, or just to say hi.
Small Snippets about Tech, AI and Machine Learning over here
AI Newsletter- https://artificialintelligencemadesimple.substack.com/
My grandma’s favorite Tech Newsletter- https://codinginterviewsmadesimple.substack.com/
Check out my other articles on Medium. : https://rb.gy/zn1aiu
My YouTube: https://rb.gy/88iwdd
Reach out to me on LinkedIn. Let’s connect: https://rb.gy/m5ok2y
My Instagram: https://rb.gy/gmvuy9
My Twitter: https://twitter.com/Machine01776819