# Decoupled Learning and Selection in Slate Recommendation for Privacy and Stability Under Noisy Scores

arXiv: https://arxiv.org/abs/2609.29453 · HTML: https://mlsanity.com/paper/2609.29453/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 26 references that are in the map.
- First posted on arXiv 24 Sep 2026.

## Builds on (most cited first, 9 of 9)

- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [Fairness of Exposure in Rankings](https://mlsanity.com/paper/1802.07281/) (2018)
- [Fairness-Aware Ranking in Search & Recommendation Systems with Application to LinkedIn Talent Search](https://mlsanity.com/paper/1905.01989/) (2019)
- [Differentially Private Contextual Linear Bandits](https://mlsanity.com/paper/1810.00068/) (2018)
- [Rank List Sensitivity of Recommender Systems to Interaction Perturbations](https://mlsanity.com/paper/2201.12686/) (2022)
- [Advancing privacy in learning analytics using differential privacy](https://mlsanity.com/paper/2501.01786/) (2025)
- [Impression-Aware Recommender Systems](https://mlsanity.com/paper/2308.07857/) (2023)

## Abstract (arXiv)

We formalize slate recommendation as a randomized score learner followed by deterministic selection. First, an appropriately scoped differential-privacy guarantee passes through selection and its audit trace by post-processing. End-to-end privacy holds only when selector inputs are public or independent, previous private outputs, or separately privacy-accounted; fixing raw state or candidate information instead yields only a conditional guarantee. Second, we derive a logged margin certificate: bounded score-induced objective movement below half the smallest greedy decision margin guarantees that the ordered slate is unchanged. Controlled fixed-margin tests show near-linear exponent scaling, with an empirical slope of -0.220 (95% CI [-0.231,-0.210]) against the independent-noise reference -1/4. Real-anchor experiments on OULAD, MovieLens-25M, and Amazon Musical Instruments show that greater anchor weight reduces score-noise-induced ranking churn. OULAD and EdNet certificate checks validate the implementation of the logged inequality, while closed-loop simulations show bounded target drift and setting-dependent downstream utility. The contribution is therefore a privacy-scope contract and a certifiable score-to-slate stability mechanism, not a universal utility claim.