Signals 4 · free daily AI digest

Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics

Paper recorded by Signals 4 on 2026-09-08 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-08 on arXiv · recorded by Signals 4 on 2026-09-09

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

Abstract

Curriculum learning is governed by several coupled design choices---how difficulty is defined, how examples are ordered, how much exposure each level receives, and how quickly training moves across levels---making it hard to isolate what actually helps. We present Wasserstein curriculum paths, a simple transport-based framework that decouples these factors by representing curricula as trajectories

Read on arXiv →

#99 most recent of 215 cs.LG papers we have recorded · ↑ newer: When Does Scale-Invariant Optimization Become Unstable? An Exact Sched · ↓ older: Multi-Task Learning for Sparsely-Labeled Time Series: A Case Study on
Cite this page: Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics: the #99 most recent of 215 cs.LG papers we have recorded (as of 2026-09-08). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/curriculum-learning-as-transport-understanding-curricula-with-wasserstein-geodes.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