Papers › Matrix Compression via Randomized Low Rank and Low Precision Factorization

Matrix Compression via Randomized Low Rank and Low Precision Factorization

17 Oct 2023NeurIPS 2023 11arXiv:2310.11028archive 2025-07-28

Rajarshi Saha, Varun Srivastava, Mert Pilanci

Matrices are exceptionally useful in various fields of study as they provide a convenient framework to organize and manipulate data in a structured manner. However, modern matrices can involve billions of elements, making their storage and processing quite demanding in terms of computational resources and memory usage. Although prohibitively large, such matrices are often approximately low rank. We propose an algorithm that exploits this structure to obtain a low rank decomposition of any matrix A as A ≈LR, where L and R are the low rank factors. The total number of elements in L and R can be significantly less than that in A. Furthermore, the entries of L and R are quantized to low precision formats – compressing A by giving us a low rank and low precision factorization. Our algorithm first computes an approximate basis of the range space of A by randomly sketching its columns, followed by a quantization of the vectors constituting this basis. It then computes approximate projections of the columns of A onto this quantized basis. We derive upper bounds on the approximation error of our algorithm, and analyze the impact of target rank and quantization bit-budget. The tradeoff between compression ratio and approximation accuracy allows for flexibility in choosing these parameters based on specific application requirements. We empirically demonstrate the efficacy of our algorithm in image compression, nearest neighbor classification of image and text embeddings, and compressing the layers of LlaMa-7b. Our results illustrate that we can achieve compression ratios as aggressive as one bit per matrix coordinate, all while surpassing or maintaining the performance of traditional compression techniques.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 13 samples from 1 repository, 12 ran (12 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

pilancilab/matrix-compressor officialmentioned in papermentioned on GitHubpytorchSyntology: 13 samples, 12 ran (0 instrument-failed) · 1 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

13 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

12ran
1unverified

Licence: 13 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from pilancilab/matrix-compressor. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

pilancilab/matrix-compressor official repository 13 samples, 12 ran (0 instrument-failed) · 1 unverified
batch_cutmix src/hlb/utils.py ran no licence file found · pointer only · 6547238fd017af95 · report fixture: signature inputs:Tensor/4/float, targets:Tensor/1/float, patch_size:int/0/int · returned tuple · level recorded 2026-09-24
batch_normalize_images src/hlb/utils.py ran fingerprinted no licence file found · pointer only · 80750b7d0796de5a · report fixture: signature input_images:Tensor/4/float, mean:Tensor/1/float, std:Tensor/1/float · returned Tensor · level recorded 2026-09-24
extract_embeddings in 9 papers src/mnet/utils.py ran no licence file found · pointer only · 246fd19433ca457a · report fixture: signature dataloader:list/1/tuple, model:callable/0/? · returned tuple · level recorded 2026-09-24
get_patches in 9 papers src/hlb/nn/builder.py ran no licence file found · pointer only · d7186cdef9b0b0a5 · report fixture: signature x:Tensor/4/float, patch_shape:tuple/1/int, dtype:dtype/0/? · returned Tensor · level recorded 2026-09-24
get_whitening_parameters src/hlb/nn/builder.py ran fingerprinted no licence file found · pointer only · 1dd822fea50fd772 · report fixture: signature patches:Tensor/4/float · returned tuple · level recorded 2026-09-24
init_whitening_conv src/hlb/nn/builder.py ran no licence file found · pointer only · 5d8137fb7e543a54 · report fixture: signature layer:callable/0/?, train_set:Tensor/4/float, num_examples:int/0/int, previous_block_data:NoneType/0/?, pad_amount:int/0/int, freeze:bool/0/bool, whiten_splits:NoneType/0/? · returned Tensor · level recorded 2026-09-24
make_random_square_masks src/hlb/utils.py ran no licence file found · pointer only · b82ee0719ccaa6a5 · report fixture: signature inputs:Tensor/4/float, mask_size:int/0/int · returned Tensor · level recorded 2026-09-24
maximum_output_rank src/lplr/utils.py ran no licence file found · pointer only · 6b7a27a3ef3f576d · report fixture: signature compression_ratio:float/0/float, b1:int/0/int, b2:int/0/int, b_nq:float/0/float, input_shape:tuple/1/int · returned int · level recorded 2026-09-24
normalize_and_shift_wrt_inner_prod src/lplr/utils.py ran fingerprinted no licence file found · pointer only · a81fb3527c31a385 · report fixture: signature X:Tensor/2/float, Y:Tensor/2/float · returned Tensor · level recorded 2026-09-24
relative_tensor_error src/shared/error.py ran fingerprinted no licence file found · pointer only · bab310a37b20a439 · report fixture: signature target:Tensor/2/float, estimated:Tensor/2/float, ord:NoneType/0/? · returned Tensor · level recorded 2026-09-24
sparse_jl_transform src/lplr/utils.py ran fingerprinted no licence file found · pointer only · 70c7d56564a6351e · report fixture: signature d:int/0/int, m:int/0/int, s:int/0/int · returned Tensor · level recorded 2026-09-24
uncompressed src/lplr/compressors.py ran fingerprinted no licence file found · pointer only · 4e097d543e13c385 · report fixture: signature X:Tensor/2/float · returned Tensor · level recorded 2026-09-24
load_mobilenet_without_head src/mnet/loader.py unverified no licence file found · pointer only · 3fcac282a5058eac · report

Tasks

Image CompressionQuantization

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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