Methods › General › Loss Functions › InfoNCE

InfoNCE

418 papers tagged archive 2025-07-28

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

InfoNCE, where NCE stands for Noise-Contrastive Estimation, is a type of contrastive loss function used for self-supervised learning.

Given a set X ={x₁, …, x_N} of N random samples containing one positive sample from p(xₜ₊ₖ|cₜ) and N − 1 negative samples from the 'proposal' distribution p(xₜ₊ₖ), we optimize:

ℒ_N = - 𝔼_X[log(fₖ(xₜ₊ₖ, cₜ))/(∑_(xⱼ∈X)fₖ(xⱼ, cₜ))]

Optimizing this loss will result in fₖ(xₜ₊ₖ, cₜ) estimating the density ratio, which is:

fₖ(xₜ₊ₖ, cₜ) ∝(p(xₜ₊ₖ|cₜ))/(p(xₜ₊ₖ))

Source: Representation Learning with Contrastive Predictive CodingSee Code · jefflai108/Contrastive-Predictive-Coding-PyTorch

In Syntology the paper matched by title, arXiv:2407.00143, held, no code (no_code; paper_only). The lookup matched a paper by its title, arXiv:2407.00143, not a method entry. The archive credits no paper with an arXiv id for this method, so whether this answer (about arXiv:2407.00143) is about this method is not checked; it does not say whether this method has code.

Origin paper of this answer, arXiv:2407.00143 (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: 418 · with a code link: 217 · where Syntology ran a sample: 89 (72 with a run with no instrument failure, 17 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (89 of 418 tagged: 72 with a run with no instrument failure, 17 where every run was a failure of Syntology's instrument)

Page 1 of 5: papers 1 to 100 of 418, 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 343 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
Contrastive Learning172
Self-Supervised Learning110
Representation Learning104
Data Augmentation28
Retrieval27
Image Classification26
Transfer Learning25
image-classification18
Object Detection17
Semantic Segmentation17
Linear evaluation16
object-detection16
Self-Supervised Image Classification15
Classification13
General Classification11
Language Modelling11
Speech Recognition10
speech-recognition10
Automatic Speech Recognition9
Clustering9

Usage over time archive 2025-07-28

Papers per year tagged with InfoNCE: 2018 to 2025, peak 97 97 0 2018: 2 papers 2018 2019: 10 papers 2019 2020: 52 papers 2020 2021: 97 papers 2021 2022: 89 papers 2022 2023: 74 papers 2023 2024: 64 papers 2024 2025: 30 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (418 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

Loss Functions

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