R.I.P.
π»
Ghosted
Random Reshuffling Dominates Stochastic Gradient Descent
June 30, 2026 Β· Grace Period Β· π COLT 2026
Authors
Zijian Liu
arXiv ID
2606.32005
Category
math.OC: Optimization & Control
Cross-listed
cs.LG,
stat.ML
Citations
0
Venue
COLT 2026
Abstract
Stochastic Gradient Descent ($\textsf{SGD}$) is one of the most classical optimization algorithms with favorable theoretical guarantees, yet the practical implementation of $\textsf{SGD}$ differs subtly from its well-known form and is often referred to as Shuffling Stochastic Gradient Descent ($\textsf{Shuffling SGD}$). A particularly popular strategy in $\textsf{Shuffling SGD}$ is Random Reshuffling ($\textsf{RR}$), which has achieved great empirical success across numerous experiments. Despite its strong performance, $\textsf{RR}$ has long been considered a heuristic due to a lack of theoretical support. Over the last decade, people have finally established provable convergence rates for $\textsf{RR}$, thus justifying its observed superiority. However, for smooth convex optimization, two clouds over the convergence theory of $\textsf{RR}$ remain to this day. More precisely, according to the current theory, $\textsf{Shuffling SGD}$ under $\textsf{RR}$ converges only when the stepsize is smaller than a threshold proportional to $1/n$, where $n$ is the number of summands in the objective (or the number of data points). Consequently, the optimally tuned theoretical rate of $\textsf{Shuffling SGD}$ under $\textsf{RR}$ is strictly worse than that of $\textsf{SGD}$ when the number of epochs is smaller than another threshold proportional to $n$. These two restrictions heavily limit the applicability of existing theories and leave a critical mismatch with practice. In this work, for the first time, we prove that $\textsf{RR}$ dominates $\textsf{SGD}$ in smooth convex optimization under any reasonable stepsize after any finite number of epochs, thereby addressing a longstanding open question.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Optimization & Control
R.I.P.
π»
Ghosted
Local SGD Converges Fast and Communicates Little
R.I.P.
π»
Ghosted
On Lazy Training in Differentiable Programming
π
π
The Cartographer
A Review on Bilevel Optimization: From Classical to Evolutionary Approaches and Applications
R.I.P.
π»
Ghosted
Learned Primal-dual Reconstruction
R.I.P.
π»
Ghosted