tutorials
Research tutorials by Bishnu Bhusal on differential privacy verification, private in-context learning, and private optimization, with worked examples and links to the original papers.
Research explanations and worked examples, with links to the original papers and implementations. Each tutorial separates the teaching example from the paper’s results and states the assumptions that matter.
Subscribe via RSS · Explore the research guide
-
DipApprox: verifying differential privacy with Gaussian noise
Inside our CCS 2025 paper: how DipApprox uses certified integration, Gaussian tail bounds, and probability intervals to prove or refute differential privacy.
-
Privacy-preserving in-context learning: synthetic demonstrations and privacy budgets
How clipped logits, public context, and private token sampling can produce reusable ICL demonstrations, with careful accounting for privacy boundaries.
-
Why gradient clipping changes Muon's geometry: a two-dimensional example
A runnable Python example shows how unequal per-example clipping rotates a matrix's polar factor, motivating the DP-Muon-IM preprint.