The authors designed PrivTab as a transformer-based model with a differentially private multi-head cross-attention (DP-MHCA) layer that bounds the influence of each individual record.
PrivTab embeds differential privacy into a tabular foundation model for fast private classification
The model pretrains to produce compact differentially private summaries via a single forward pass, outperforming traditional private methods under moderate-to-strong privacy.
Top University
Talal Alrawajfeh · Cristiana Diaconu · Ossi Räisä · Sebastian Rodriguez Beltran · Yuan He · John Bronskill · +2 more
University of Helsinki · University of Cambridge · CISPA Helmholtz Center for Information Security
Research Digest··2 min read
Alrawajfeh et al.
Why this paper
From CISPA Helmholtz Center for Information Security and 2 others
In one line
PrivTab delivers differentially private tabular classification in one forward pass, outperforming private linear and neural baselines under moderate-to-strong privacy while fitting 10,000 times faster.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors (2 noted)
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§