Methods › General › Stochastic Optimization › SGD

Stochastic Gradient Descent

SGD

introduced 1951 2,021 papers tagged archive 2025-07-28

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using all available data. That is for weights w and a loss function L we have:

wₜ₊₁ = wₜ - η∇̂_wL(wₜ)

Where η is a learning rate. SGD reduces redundancy compared to batch gradient descent - which recomputes gradients for similar examples before each parameter update - so it is usually much faster.

(Image Source: here)

See Code · pytorch/pytorch

In Syntology name not resolved (unknown_name). Other entries in Syntology's graph carry this name; the lookup's note below says what that means for this name.

From the lookup's note on this name: Nothing resolved to this name, but 5 entries in Syntology's graph carry it; the most cited is the paper “SGD: General Analysis and Improved Rates” (arXiv:1901.09401, 488 citations). The lookup reads this as a miss of its own name matching, not a gap in coverage: the graph holds the name under a spelling the lookup did not reach. For arXiv:1901.09401: ask the lookup about it.

Near matches the lookup lists, each a different name: ∞-SGD; SGD-KV; INO-SGD; FBi-SGD; badr-sgd; FO-DP-SGD; DP-SGD-RC; SMA-DP-SGD.

Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.

Papers archive 2025-07-28

archive papers tagged: 2,021 · with a code link: 591 · where Syntology ran a sample: 192 (160 with a run with no instrument failure, 32 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (192 of 2,021 tagged: 160 with a run with no instrument failure, 32 where every run was a failure of Syntology's instrument)

Page 1 of 21: papers 1 to 100 of 2,021, newest first by the archive's date (ties by slug), in archive order.

Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.

Tasks archive 2025-07-28

20 shown of 362 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Stochastic Optimization149
Image Classification122
Federated Learning110
image-classification102
BIG-bench Machine Learning66
Deep Learning66
regression59
Quantization55
Language Modelling52
GPU49
Generalization Bounds45
Language Modeling44
Distributed Optimization39
General Classification39
Computational Efficiency27
Data Augmentation27
Second-order methods25
Privacy Preserving24
Learning Theory23
Speech Recognition23

Usage over time archive 2025-07-28

Papers per year tagged with SGD: 2011 to 2025, peak 337 337 0 2011: 3 papers 2011 2012: 2 papers 2013: 10 papers 2013 2014: 10 papers 2015: 25 papers 2015 2016: 41 papers 2017: 75 papers 2017 2018: 152 papers 2019: 218 papers 2019 2020: 291 papers 2021: 337 papers 2021 2022: 226 papers 2023: 270 papers 2023 2024: 244 papers 2025: 116 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (2,020 dated, 1 undated not charted). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Stochastic Optimization

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections