No article text was recorded for this version.
training horizon dependency
Version 1 of 122 papers24 conceptsfrozen 2026-10-04
Papers in this version
- Blog: How far away are truly hyperparameter-free learning algorithms?
- Blog: Celo2: Towards Learned Optimization Free Lunch
- Blog: Power Scheduler: A Batch Size and Token Number Agnostic Learning Rate Scheduler
- Blog: WSM: Decay-Free Learning Rate Schedule via Checkpoint Merging for LLM Pre-training
- Blog: Adaptive Variance Reduction for Stochastic Optimization under Weaker Assumptions
- Blog: Gradient Descent Averaging and Primal-dual Averaging for Strongly Convex Optimization
- Blog: The Step Decay Schedule: A Near Optimal, Geometrically Decaying Learning Rate Procedure For Least Squares
- Blog: Learning to Schedule Learning rate with Graph Neural Networks
- Blog: The Road Less Scheduled
- Blog: Through the River: Understanding the Benefit of Schedule-Free Methods for Language Model Training
- Blog: On the Last-Iterate Convergence of Shuffling Gradient Methods
- Blog: The Surprising Agreement Between Convex Optimization Theory and Learning-Rate Scheduling for Large Model Training
- Blog: Stab-SGD: Noise-Adaptivity in Smooth Optimization with Stability Ratios
- Blog: Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations
- Blog: Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training
- Blog: Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
- Blog: Adaptive Stochastic Variance Reduction for Non-convex Finite-Sum Minimization
- Blog: Understanding Warmup-Stable-Decay Learning Rates: A River Valley Loss Landscape View
- Blog: Training Dynamics of the Cooldown Stage in Warmup-Stable-Decay Learning Rate Scheduler
- Blog: Amos: An Adam-style Optimizer with Adaptive Weight Decay towards Model-Oriented Scale
- Blog: Analysis of Schedule Free Non-Convex Optimization
- Blog: REX: Revisiting Budgeted Training with an Improved Schedule
History & citation
Current · Permanent link
- This direction split from Schedule-Free Optimization for Flexible Training Horizons (64% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_training_horizon_dependency_eeb8008d,
title = {training horizon dependency},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/training-horizon-dependency/rdr_eeb8008d8ee6493a82b0a7fc0593332e},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 1, frozen text):
@misc{lacuna_training_horizon_dependency_eeb8008d_v1,
title = {training horizon dependency},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/training-horizon-dependency/rdr_eeb8008d8ee6493a82b0a7fc0593332e/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}