Backend
Distributionally Robust Risk Evaluation with Low-Rank Perturbations
arXiv stat.ML0 views
arXiv:2611.17447v1 Announce Type: cross Abstract: We study distributionally robust risk under perturbations restricted to low-rank matrices whose row and column spaces are shared across observations, focusing on how much worst-case risk this restriction sacrifices and how to compute the restricted worst case. Under moment or Wasserstein budgets, we derive quantitative bounds on the gap between the unrestricted and low-rank worst-case risks, characterize when the gap vanishes, and identify the smallest lossless row and column dimensions. To generate worst-case perturbations, we apply projected gradient ascent to a curvature-corrected concave problem and derive computable risk certificates. We establish global sublinear convergence for the corrected objective and linear convergence to a worst-case distribution when the budget constraint is sufficiently active.
Source:https://arxiv.org/abs/2611.17447PreviousElliptic Sampling: Optimal Interpolant Schedules for Gaussians and BeyondNextSURFACE: Preference Distribution Matching on Pareto Fronts via Scalarization Steering
Related posts
LLVM's Old AMD R600 Compiler Backend Slated For Removal
0Phoronix

Give your AI agent real Google rankings with a self-hosted MCP server
0dev.to
Fix: GitHub Pull Request Merges Timing Out
0dev.to

Stop Trusting Your Cache Hit Rate. 7 Tests Show Why
0dev.to

Your AI agent's public URL is an open wallet
0dev.to

Extending csperf: Plugins Without Breaking Honesty
0dev.to