MétaCan
Menu
Back to cohort
Record W7133000605

Recovering Utility in LDP Schemes by Training with Noise^2

2024· dissertation· W7133000605 on OpenAlexaff
Kexin Li

Bibliographic record

VenueTSpace · 2024
Typedissertation
Language
FieldComputer Science
TopicAdversarial Robustness in Machine Learning
Canadian institutionsUniversity of Toronto
Fundersnot available
KeywordsInferenceNoise (video)Key (lock)Scheme (mathematics)Training setSet (abstract data type)Training (meteorology)Robustness (evolution)
DOInot available

Abstract

fetched live from OpenAlex

The adoption of large cloud-based models for inference has been hampered by concerns about the privacy leakage of end-user data. One method to mitigate this leakage is to add local differentially private noise to queries before sending them to the cloud, but this degrades utility as a side effect. Our key insight is that knowledge available in the noisy labels returned from performing inference on noisy inputs can be aggregated and used to recover the correct labels. We implement this insight in LDPKiT, which stands for Local Differentially-Private and Utility-Preserving Inference via Knowledge Transfer. LDPKiT uses the noisy labels returned from querying a set of noised inputs to train a local model (noiseˆ2 ), which is then used to perform inference on the original set of inputs. Our experiments on CIFAR-10, Fashion-MNIST, SVHN, and CARER NLP datasets demonstrate that LDPKiT can improve utility without compromising privacy. For instance, on CIFAR-10, compared to a standard ϵ-LDP scheme applying the Laplacian noise on each data sample with ϵ = 15, which provides a weak privacy guarantee, LDPKiT can achieve similar accuracy with ϵ = 7, offering an enhanced privacy guarantee. Moreover, the benefits of using LDPKiT increase at higher, more privacy-protective noise levels. For Fashion-MNIST and CARER, LDPKiT’s accuracy on the sensitive dataset with ϵ = 7 not only exceeds the average accuracy of the standard ϵ-LDP scheme with ϵ = 7 by roughly 20% and 9% but also outperforms the standard ϵ-LDP scheme with ϵ = 15, a scenario with less noise and minimal privacy protection. We also perform Zest distance measurements to demonstrate that the type of distillation performed by LDPKiT is different from a model extraction attack.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.010
metaresearch head score (Gemma)0.038
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.010
Threshold uncertainty score0.055

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0100.038
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0020.001
Bibliometrics0.0010.002
Science and technology studies0.0020.004
Scholarly communication0.0030.008
Open science0.0050.007
Research integrity0.0030.005
Insufficient payload (model declined to judge)0.0030.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.027
GPT teacher head0.333
Teacher spread0.306 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2024
Admission routes1
Has abstractyes

Explore more

Same venueTSpaceSame topicAdversarial Robustness in Machine LearningFrench-language works237,207