🧠机器之心•Stalecollected in 12m
PIL: Linear Proxies for Unlearnable Samples
#research#pil#iclr-2026#unlearnable-examples#data-privacypil
💡100x faster unlearnable examples via linear proxies – essential data privacy tool (ICLR 2026)
⚡ 30-Second TL;DR
What Changed
Replaces DNN proxies with linear models for PGD optimization
Why It Matters
Democratizes unlearnable examples for photographers/users, enabling practical data protection against model scraping at low cost.
What To Do Next
Run PIL from https://github.com/jinlinll/pil on your images to test unlearnability vs ResNet.
Who should care:Researchers & Academics
Key Points
- •Replaces DNN proxies with linear models for PGD optimization
- •Key insight: unlearnable samples boost model linearity (FGSM metric)
- •15+ GPU hours (REM) → minutes; scales to high-res images
- •Open-source code protects data privacy without heavy compute
📰
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: 机器之心 ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
Weekly AI briefing
One email a week. Unsubscribe anytime.