Thread by @TheSequenceAI

Deep dive into "Scalable Private Learning with PATE" by @NicolasPapernot @_kunal_talwar_ @UlfarEr Shuang Song, Ilya Mironov, Ananth Raghunathan

It presents a Private Aggregation of Teacher Ensembles (PATE) method to ensure privacy in training datasets
Thread👇🏼 🔎

Imagine that two different models, trained on two different datasets produce similar outputs

Then, their decision does not reveal information about any single training example

And this is another way to say it ensures the privacy of the training data
2/⬇️

PATE uses a perturbation technique that structures the learning process using an ensemble of teacher models communicating their knowledge to a student model
3/⬇️

PATE fundamental components:

+Teacher Models
+Aggregation Mechanisms: evaluate when the teacher models achieve consensus on output
+Student Models
4/⬇️

PATE = one of the most influential models in private machine learning, having been adopted by several frameworks and tools

Link: arxiv.org/pdf/1802.08908…
Github: github.com/tensorflow/pri…
5/⬇️

TheSequence Edge covers:
+ML concept you should learn
+Review of an impactful research paper
+New ML framework or platform and how you can use it
thesequence.substack.com/subscribe
6/6