יום שלישי, 15 בספטמבר 2026 LIVE
AI־INFO

כתבה arXiv cs.LG ·

למידת תוכנית כתחבורה

Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics
חוקרים פיתחו מסגרת חדשה ללמידת תוכנית, המייצגת תוכניות כמסלולים של הפצות אימון. המחקר בוחן את השפעת הסדר, חשיפה וקצב על ביצועים.
תקציר מקורי באנגליתarXiv:2609.09099v1 Announce Type: new 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 of training distributions over discrete difficulty levels. Across a calibrated synthetic suite with 12 tasks and 33 difficulty axes, we use this framework to isolate the effects of ordering, matched exposure, endpoint smoothness, and pacing under fixed training budgets. We find that curriculum effects are strongly context-dependent: no single strate
קרא במקור המקורי