the open
    research
    Contribute
    Back to the reading lists

    Computer vision & multimodal· 2018

    Representation Learning with Contrastive Predictive Coding

    Aaron van den Oord (first author) · 2018

    Read the original

    arXiv 1807.03748OpenAlex W4297808394

    On the reading lists

    Why it matters

    Introduced CPC and InfoNCE, foundational objectives for modern contrastive self-supervised learning.

    Condensed from the expert lists' notes.

    Reading it

    Difficulty: Advanced

    Read this first: probability (mutual information); neural network basics (backpropagation).

    The record

    First released
    2018 (from arXiv 1807.03748)
    Citations
    14,862 citations, as counted by Papers With Code (CC BY-SA 4.0)
    Named by
    both expert lists
    Also relevant to
    Language, speech & LLMs
    Score
    0.633 · computed rank 123 · consensus 0.9, must-read 1, citations percentile 0.753 (2010s · vision), awards 0, existing lists 0. Method

    Editor's notes

    • Readability check: pass: Introduces the InfoNCE loss used across self-supervised learning. (AI draft (Claude), not yet reviewed, 2026-10-29)

    Data sources

    Catalogue data from OpenAlex (CC0), Crossref (CC0 metadata; no abstracts are shown), arXiv, and Papers With Code (CC BY-SA 4.0). Each count names its source. See the method.