← Back to all news
arXiv Machine Learning By Hanqing Liu, Jianjun Cao, Yuanze Li, Zijian Zhou

Grokking or Glitching? How Low-Precision Drives Slingshot Loss Spikes

Read the original on arXiv Machine Learning →

Deep neural networks can exhibit periodic loss spikes known as the "Slingshot Mechanism" during long‑term training. This paper shows that such spikes arise from finite‑precision floating‑point arithmetic: when the correct‑class logit becomes much larger than the others, the gradient for that class is rounded to zero while others remain nonzero, breaking the zero‑sum constraint and creating a positive feedback loop that inflates feature and classifier norms. The authors term this Numerical Feature Inflation (NFI) and demonstrate that it explains the rapid norm growth, reappearance of gradients, and loss spikes, while also showing that partial absorption can still cause hidden norm growth without visible spikes.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Aug 4

GradientStabilizer:Fix the Norm, Not the Gradient

arXiv:2502. 17055v5 Announce Type: replace Abstract: Training instability in modern deep learning systems is frequently triggered by rare but extreme gradient-norm spikes, which can induce oversized parameter updates, corrupt optimizer state, and lead to slow recovery or divergence.

By Tianjin Huang, Zhangyang Wang, Haotian Hu, Zhenyu Zhang, Gaojie Jin, Xiang Li, Li Shen, Jiaxing Shang, Tianlong Chen, Ke Li, Lu Liu, Qingsong Wen, Shiwei Liu
llmsreinforcement-learningefficiency
More like this →
arXiv Machine Learning
Sep 11

ExpTest: Loss-Curve Hypothesis Testing for Autonomous Learning-Rate Selection in Deep Neural Networks

ExpTest is an autonomous learning‑rate controller that uses the training loss curve as an online signal to perform sequential statistical tests on theoretically motivated windows, detecting convergent behavior and triggering learning‑rate reductions. It combines a covariance‑based initial learning‑rate estimate, curvature‑motivated window sizing, and a two‑phase test‑driven decay, relying on the approximately exponential decay predicted under linearized network dynamics. Experiments on regression, classification, forecasting, and natural‑language tasks across various architectures show that ExpTest achieves competitive performance compared to hand‑tuned SGD baselines and recent learning‑rate‑free methods, without requiring manual initial learning‑rate selection or predefined scheduling.

By Zan Chaudhry, Naoko Mizuno
llmsagents
More like this →