← Back to all news
arXiv Machine Learning By Zhongjing Gu, Chenyang Huang, Yufa Feng, Chong He, Qinxu Ding, Yiming Cui

VCMM: Variance-Calibrated Momentum for Multimodal Learning

Read the original on arXiv Machine Learning →

VCMM: Variance-Calibrated Momentum for Multimodal Learning proposes a new optimizer that adapts momentum based on modality-specific gradient dynamics. It estimates minibatch noise and temporal drift online, using a Kalman-inspired controller to set modality-specific momentum and applies bias correction for the first moment. Experiments on four multimodal benchmarks show consistent improvements with modest training overhead.

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 18

Adaptive Optimization via Momentum on Variance-Normalized Gradients

arXiv:2602. 10204v2 Announce Type: replace Abstract: We introduce MVN-Grad (Momentum on Variance-Normalized Gradients), an Adam-style optimizer that improves stability and performance by combining two complementary ideas: variance-based normalization and momentum applied after normalization.

By Francisco Patitucci, Aryan Mokhtari
llms
More like this →
arXiv Machine Learning
5d ago

Gradient-Momentum Coupling: A Parameter-Space Proxy for Learning Progress

The paper introduces Gradient‑Momentum Coupling (GMC), a method that quantifies learning progress by measuring how strongly a sample influences changes in the parameter space, using the normalized absolute product of its gradient and the momentum of previous gradients. GMC filters out noise by accumulating consistent directions of change while canceling random fluctuations, leading to a more uniform prioritization across tasks with varying noise levels and better ranking of learnable tasks by improvement speed. Experiments on MiniGrid MultiRoom tasks show that replacing prediction error with GMC in the Intrinsic Curiosity Module restores exploration capabilities that were lost to unpredictable observations.

By Samuel Blad, Martin L\"angkvist, Amy Loutfi
agents
More like this →