EconBase
← All papers

PPI-SVRG: Unifying Prediction-Powered Inference and Variance Reduction for Semi-Supervised Optimization

Ruicheng Ao, Hongyu Chen, Haoyang Liu, David Simchi-Levi, Will Wei Sun

arXiv 29 Jan 2026 · Machine Learning

arXiv:2601.21470 · PDF · DOI · OpenAlex · Extracted main text

Abstract

We study semi-supervised stochastic optimization when labeled data is scarce but predictions from pre-trained models are available. PPI and SVRG both reduce variance through control variates -- PPI uses predictions, SVRG uses reference gradients. We show they are mathematically equivalent and develop PPI-SVRG, which combines both. Our convergence bound decomposes into the standard SVRG rate plus an error floor from prediction uncertainty. The rate depends only on loss geometry; predictions affect only the neighborhood size. When predictions are perfect, we recover SVRG exactly. When predictions degrade, convergence remains stable but reaches a larger neighborhood. Experiments confirm the theory: PPI-SVRG reduces MSE by 43--52% under label scarcity on mean estimation benchmarks and improves test accuracy by 2.7--2.9 percentage points on MNIST with only 10% labeled data.

Citation extraction

21
references
38
in-text mentions
21
distinct cited
1
self-citations
5,485
main-text words

appendix boundary found by appendix_command · 49% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Rie Johnson and Tong Zhang (2013) Accelerating stochastic gradient descent using predictive variance reduction1.00054100%
2Anastasios N Angelopoulos, Stephen Bates, Clara Fannjiang, Michael I… (2023) Prediction-powered inference0.9285580%
3Zeyuan Allen-Zhu and Yang Yuan (2016) Improved svrg for non-strongly-convex or sum-of-non-convex objectives0.6936333%
4Frank Schneider, Lukas Balles, and Philipp Hennig (2019) Deepobs: A deep learning optimizer benchmark suite0.5114225%
5Liudmila Prokhorenkova, Gleb Gusev, Aleksandr Vorobev, Anna Veronika… (2018) Catboost: unbiased boosting with categorical features0.5112250%
6Anastasios N Angelopoulos, John C Duchi, and Tijana Zrnic (2023) Ppi++: Efficient prediction-powered inference0.40511100%
7Ruicheng Ao, Hongyu Chen, and David Simchi-Levi (2024) Prediction-guided active experiments self0.40511100%
8Aaron Defazio, Francis Bach, and Simon Lacoste-Julien (2014) Saga: A fast incremental gradient method with support for non-strongly convex composite objectives0.40511100%
9Cong Fang, Chris Junchi Li, Zhouchen Lin, and Tong Zhang (2018) Spider: Near-optimal non-convex optimization via stochastic path-integrated differential estimator0.40511100%
10Neal Jean, Marshall Burke, Michael Xie, W Matthew Davis, David B Lob… (2016) Combining satellite imagery and machine learning to predict poverty0.40511100%

Showing the top 10 of 21 scored citations.