Methods › General › Large Batch Optimization › 1-bit Adam

1-bit Adam

40 papers tagged archive 2025-07-28

Introduced by Hanlin Tang et al. in 1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed

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

1-bit Adam is a stochastic optimization technique that is a variant of ADAM with error-compensated 1-bit compression, based on finding that Adam's variance term becomes stable at an early stage. First vanilla Adam is used for a few epochs as a warm-up. After the warm-up stage, the compression stage starts and we stop updating the variance term v and use it as a fixed precondition. At the compression stage, we communicate based on the momentum applied with error-compensated 1-bit compression. The momentums are quantized into 1-bit representation (the sign of each element). Accompanying the vector, a scaling factor is computed as (magnitude of compensated gradient)/(magnitude of quantized gradient). This scaling factor ensures that the compressed momentum has the same magnitude as the uncompressed momentum. This 1-bit compression could reduce the communication cost by 97 % and 94 % compared to the original float 32 and float 16 training, respectively.

PaperSource

In Syntology repository known, no verified implementation (declared_code; repositories: 2 listed of 2 in the graph, strongest link: official). The lookup matched a paper by its title, arXiv:2102.02888, not a method entry. This answer is about arXiv:2102.02888, the paper the archive credits with this method.

Origin paper of this answer, arXiv:2102.02888 (its page on this site · its page on Syntology).

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: 40 · with a code link: 33 · where Syntology ran a sample: 12 (12 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (12 of 40 tagged: 12 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument)

Page 1 of 1: papers 1 to 40 of 40, 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 96 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
Semantic Segmentation5
Image Generation4
2D Object Detection3
Image Classification3
Object Detection3
Retrieval3
object-detection3
2D Human Pose Estimation2
Deep Learning2
General Classification2
Image Retrieval2
Language Modeling2
Language Modelling2
Large Language Model2
Multi-task Language Understanding2
Object2
Text to Image Generation2
Text-to-Image Generation2
Video Generation2
image-classification2

Usage over time archive 2025-07-28

Papers per year tagged with 1-bit Adam: 1986 to 2025, peak 7 7 0 1986: 1 paper 1986 1987: 0 papers 1988: 0 papers 1989: 0 papers 1990: 0 papers 1991: 0 papers 1991 1992: 0 papers 1993: 0 papers 1994: 0 papers 1995: 0 papers 1996: 0 papers 1996 1997: 0 papers 1998: 0 papers 1999: 0 papers 2000: 0 papers 2001: 0 papers 2001 2002: 0 papers 2003: 0 papers 2004: 0 papers 2005: 0 papers 2006: 0 papers 2006 2007: 0 papers 2008: 0 papers 2009: 0 papers 2010: 0 papers 2011: 0 papers 2011 2012: 0 papers 2013: 0 papers 2014: 0 papers 2015: 1 paper 2016: 1 paper 2016 2017: 4 papers 2018: 4 papers 2019: 5 papers 2020: 3 papers 2021: 7 papers 2021 2022: 3 papers 2023: 5 papers 2024: 4 papers 2025: 2 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (40 dated). 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

Large Batch OptimizationStochastic 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