Pith. sign in

REVIEW 3 cited by

The Road Less Scheduled

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.15682 v4 pith:FZZECVVP submitted 2024-05-24 cs.LG cs.AImath.OCstat.ML

Aaron Defazio , Xingyu Alice Yang , Harsh Mehta , Konstantin Mishchenko , Ahmed Khaled , Ashok Cutkosky This is my paper
classification cs.LGcs.AImath.OCstat.ML
keywords scheduleslearningproblemsapproachmethodrateschedule-freestopping
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Existing learning rate schedules that do not require specification of the optimization stopping step T are greatly out-performed by learning rate schedules that depend on T. We propose an approach that avoids the need for this stopping time by eschewing the use of schedules entirely, while exhibiting state-of-the-art performance compared to schedules across a wide family of problems ranging from convex problems to large-scale deep learning problems. Our Schedule-Free approach introduces no additional hyper-parameters over standard optimizers with momentum. Our method is a direct consequence of a new theory we develop that unifies scheduling and iterate averaging. An open source implementation of our method is available at https://github.com/facebookresearch/schedule_free. Schedule-Free AdamW is the core algorithm behind our winning entry to the MLCommons 2024 AlgoPerf Algorithmic Efficiency Challenge Self-Tuning track.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Causal Optimizer Interaction Calculus: Hidden Geometric Relaxation and Identifiable Interventions

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Under fixed innovation coupling, finite-horizon optimizers admit minimal pathwise realizations and incidence-identifiable Möbius effects, with a five-term readout transfer from hidden relaxation and a closed reduced-v...

  2. Anomaly Object Segmentation with Vision-Language Models for Steel Scrap Recycling

    cs.CV 2025-06 conditional novelty 5.0 of 10

    Supervised finetuning of a CLIP image encoder with multi-scale features and text prompts detects anomaly objects in steel scrap at 28.6% pixel-level average precision, outperforming tested baselines on a private dataset.

  3. Analysis of Schedule-Free Nonconvex Optimization

    cs.LG 2025-08 conditional novelty 4.0 of 10

    A Lyapunov framework yields O(1/log T) and O(log T/T) gradient-norm rates for Schedule-Free on smooth nonconvex objectives, with the faster rate depending on an unproven assumption.

Pith tools