Methods › General › Negative Sampling › Self-Adversarial Negative Sampling

Self-Adversarial Negative Sampling

30 papers tagged archive 2025-07-28

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

Self-Adversarial Negative Sampling is a negative sampling technique used for methods like word embeddings and knowledge graph embeddings. The traditional negative sampling loss from word2vec for optimizing distance-based models be written as:

L = −logσ(γ− dᵣ(𝐡, 𝐭)) − ∑ⁿᵢ₌₁1/klogσ(dᵣ(𝐡^′ᵢ, 𝐭^′ᵢ) - γ)

where γ is a fixed margin, σ is the sigmoid function, and (𝐡^′ᵢ, r, 𝐭^′ᵢ) is the i-th negative triplet.

The negative sampling loss samples the negative triplets in a uniform way. Such a uniform negative sampling suffers the problem of inefficiency since many samples are obviously false as training goes on, which does not provide any meaningful information. Therefore, the authors propose an approach called self-adversarial negative sampling, which samples negative triples according to the current embedding model. Specifically, we sample negative triples from the following distribution:

p(h^′ⱼ, r, t^′ⱼ | set(hᵢ, rᵢ, tᵢ ) ) = (expαfᵣ(𝐡^′ⱼ, 𝐭^′ⱼ))/(∑ᵢ₌₁expαfᵣ(𝐡^′ᵢ, 𝐭^′ᵢ))

where α is the temperature of sampling. Moreover, since the sampling procedure may be costly, the authors treat the above probability as the weight of the negative sample. Therefore, the final negative sampling loss with self-adversarial training takes the following form:

L = −logσ(γ− dᵣ(𝐡, 𝐭)) − ∑ⁿᵢ₌₁p(h^′ᵢ, r, t^′ᵢ)logσ(dᵣ(𝐡^′ᵢ, 𝐭^′ᵢ) - γ)

Source: RotatE: Knowledge Graph Embedding by Relational Rotation...

Papers archive 2025-07-28

30 shown of 30, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 51 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
Graph Embedding20
Knowledge Graph Embedding19
Knowledge Graphs14
Link Prediction13
Knowledge Graph Completion12
Relation11
Entity Embeddings4
Knowledge Graph Embeddings4
Translation4
Representation Learning3
Triplet3
Benchmarking2
Computational Efficiency2
Decoder2
GPU2
Graph Representation Learning2
Prediction2
Reinforcement Learning (RL)2
Relational Reasoning2
Sentence2

Usage over time archive 2025-07-28

Papers per year tagged with Self-Adversarial Negative Sampling: 2019 to 2025, peak 9 9 0 2019: 4 papers 2019 2020: 2 papers 2020 2021: 7 papers 2021 2022: 9 papers 2022 2023: 2 papers 2023 2024: 4 papers 2024 2025: 2 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (30 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

Negative Sampling

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