On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

count % of corpus

5 results for “RERM”

ERM and RERM are optimal estimators for regression problems when malicious outliers corrupt the labelsmath.ST

ERM and RERM minimize error even with malicious label corruptions.

problem Malicious label corruptions in regression problems.

method Empirical Risk Minimizers (ERM) and Regularized Empirical Risk Minimizers (RERM) under a local Bernstein condition.

result The

L_2

-error rate is bounded by $r_N + AL |\cO|/N$ under the local Bernstein condition.

regression malicious outliers ERM RERM empirical risk minimization

2019-10-24abs ↗pdf ↗

On the robustness of minimum norm interpolators and regularized empirical risk minimizersmath.ST

The study analyzes robustness of estimators in linear models with adversarial errors.

problem Analyzing robustness of estimators in linear models with adversarial errors.

method Develops a general theory for minimum norm interpolating estimators and RERM in linear models without conditions on errors.

result Quantitative bound for the prediction error relating it to Rademacher complexity, norm of minimum norm interpolator of errors, and subdifferential size.

robustness estimators linear models adversarial errors minimum norm

2020-12-01abs ↗pdf ↗

Robust Empirical Risk Minimization with Tolerancecs.LG

New method reduces sample complexity for robust learning.

problem Developing simple, sample-efficient learning algorithms for robust classification.

method Tolerant empirical risk minimization (RERM) for robust learning.

result Tolerant RERM requires only O(VC(H)dlog(D/γδ)/ε^2) samples for robustness regions of diameter D.

empirical risk minimization robust learning tolerant robustness sample complexity VC classes

2022-10-02abs ↗pdf ↗

SAPPHIRE: Preconditioned Stochastic Variance Reduction for Faster Large-Scale Statistical Learningstat.ML

SAPPHIRE tackles ill-conditioned rERM problems with faster convergence.

problem Ill-conditioned objectives and non-smooth regularizers in rERM.