REVIEW 6 cited by
Optimal Linear Decay Learning Rate Schedules and Further Refinements
Not yet reviewed by Pith; the record is open.
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
read the original abstract
Learning rate schedules used in practice bear little resemblance to those recommended by theory. We close much of this theory/practice gap, and as a consequence are able to derive new problem-adaptive learning rate schedules. Our main technical contribution is a refined analysis of learning rate schedules for a wide class of optimization algorithms (including SGD). When considering only worst-case analysis, our theory predicts that the optimal choice is the linear decay schedule where the step-size is set proportional to 1 - t/T, where t is the current iteration and T is the total number of steps. To go beyond this worst-case analysis, we use the observed gradient norms to derive schedules refined for any particular task. These refined schedules exhibit learning rate warm-up and rapid learning rate annealing near the end of training. Ours is the first systematic approach to automatically yield both of these properties. We perform the most comprehensive evaluation of learning rate schedules to date, evaluating across 10 diverse deep learning problems, a series of LLMs, and a suite of logistic regression problems. We validate that overall, the linear-decay schedule outperforms all commonly used default schedules including cosine annealing. Our adaptive schedule refinement method gives further improvements.
Forward citations
Cited by 6 Pith papers
-
A Defense of the Quadratic Model
Local Taylor-expanded quadratic models reproduce a 150M-parameter LLM's validation loss for up to 10% of training late in the run, and LLM pretraining operates within a factor of 2 of a stochastic or deterministic edg...
-
WSqD: A Horizon-Free Learning Rate Schedule for Large Model Training
WSqD uses a horizon-independent shifted inverse-square-root base plus linear cooldown to get optimal last-iterate rates in convex stochastic optimization and match or beat tuned WSD on LLM pretraining with one reused ...
-
Sharp asymptotic theory for Q-learning with LDTZ learning rate and its generalization
Q-learning with PD2Z/LD2Z step sizes admits sharp non-asymptotic bounds, a tail Polyak–Ruppert CLT, and a time-uniform Gaussian approximation, establishing a best-of-both-worlds rate-and-bias tradeoff.
-
Step-Size Stability in Stochastic Optimization: A Theoretical Perspective
The stability index δ_t, the variance term in a two-term suboptimality bound, is provably smaller and saturates in the step-size cap for SPS, NGN, and proximal point, explaining their observed learning-rate robustness...
-
Why Do We Need Warm-up? A Theoretical Perspective
Under the proposed (H0,H1)-smoothness condition, gradient descent with a warm-up-style adaptive step-size provably converges faster than with any fixed step-size.
-
WSM: Decay-Free Learning Rate Schedule via Checkpoint Merging for LLM Pre-training
Checkpoint merging during constant-LR training can replace LR decay and yields improved LLM benchmark scores over Warmup-Stable-Decay.
Discussion (0). Continue with ORCID to comment.