Boosted and Differentially Private Ensembles of Decision Trees

Boosted and Differentially Private Ensembles of Decision Trees

Channel:
United States Google TechTalks
Subscribers:
348,000
Published on ● Video Link: https://www.youtube.com/watch?v=klkvl_nv_L8


Duration: 15:47
220 views
3

A Google TechTalk, 2020/7/29, presented by Richard Nock, Data61 and The Australian National University
ABSTRACT: Ensembles of decision tree (DT) classifiers are hugely popular in
both the private and non-private settings, and display a very singular
picture: while boosting and offsprings top international competitions
in the non-private setting, random forests reign supreme
when differential privacy (DP) is at stake. There is no middle ground
that would combine the rates of boosting with the randomness that DP
commands. In this talk, we summarise (i) the existence of a privacy vs boosting dilemma for top-down induction of DTs in the context of statistical decision theory, (ii) an algorithm to navigate this dilemma with the introduction of a new specifically designed proper loss and a way to tune it at training time to make the most of boosting under DP constraints, (iii) formal boosting convergence results under DP constraints and (iv) experiments comparing our approach to differentially private random forests.



Other Videos By Google TechTalks


2021-06-04Breaking the Communication-Privacy-Accuracy Trilemma
2021-06-04Cronus: Robust Knowledge Transfer for Federated Learning
2021-06-04Private Algorithms with Minimal Space
2021-06-04Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization
2021-06-04Flower: A Friendly Federated Learning Framework
2021-06-04Lions, Skunks, and Kangaroos: Geo-Distributed Learning on the Flickr-Mammal Dataset
2021-06-04Secure Federated Learning in Adversarial Environments
2021-06-04Analyzing Model Poisoning Attacks on Federated Learning at Scale
2021-06-04Federated Design of Compact and Private DNNs
2021-06-04FL Meets Wireless: More than the sum of its parts
2021-06-04Boosted and Differentially Private Ensembles of Decision Trees
2021-06-04On Heterogeneity in Federated Settings: Workshop on Federated Learning and Analytics Day 2 Keynote
2021-06-04Towards Responsible AI
2021-06-04Workshop on Federated Learning & Analytics: Pre-recorded Talks Day 1 Track 2 Q&A Privacy/Security
2021-06-04Context-Aware Local Differential Privacy
2021-06-04Semantic Adversarial Attacks for Privacy Protection
2021-06-04How Could the Mind be Storing Information?
2021-05-17Towards Reliable Machine Learning via Distributional Robustness
2021-05-10Advent of Code: Behind the Scenes
2021-04-212021 Blockly Developer Summit: Test Helpers
2021-04-212021 Blockly Developer Summit: Metrics Deep Dive