Signals 4 · free daily AI digest

Benign Loss Landscapes Can Coexist with Worst-Case Hardness

Paper recorded by Signals 4 on 2026-09-11 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-11 on arXiv · recorded by Signals 4 on 2026-09-14

Category: cs.LG · 机器学习 · first seen 2026-09-14

Abstract

Deep neural networks are expressive enough to contain worst-case targets that can be evaluated in polynomial time but cannot be learned in polynomial time by gradient descent. For practical tasks they nonetheless learn well, raising the question of what non-generic structure of real-world targets enables this. Existing surrogate models cannot pose this question because they either lack hard-to-lea

Read on arXiv →

#58 most recent of 215 cs.LG papers we have recorded · ↑ newer: CanvasAnneal: Curriculum Reinforcement Learning for Diffusion Language · ↓ older: A Unified and Constrained View of Regularization-Based Robust Reinforc
Cite this page: Benign Loss Landscapes Can Coexist with Worst-Case Hardness: the #58 most recent of 215 cs.LG papers we have recorded (as of 2026-09-11). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/benign-loss-landscapes-can-coexist-with-worst-case-hardness.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.LG papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions