Research map / Learning theory and algorithms / Sample complexity
Provable Accuracy Collapse of Embedding-Based Representations under Dimensionality Mismatch
Dionysis Arvanitakis, Vaggos Chatziafratis, Yiyuan Luo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| reduce embedding dimension d below a constant fraction of ground-truth dimension D (ablation) | triplet-loss embeddings (hinge loss, AdamW) · synthetic sphere points n=1000, m=10^6 triplets, D in {128,256,512,1024} | triplet accuracy: accuracy falls toward the 50% random baseline; d/D about 5% gives accuracy about 1/2 + 22% | Hurt |
| reduce embedding dimension d on uniformly random realizable triplets (ablation) | triplet-loss embeddings · random triplets n=4000, m=10^6 | triplet accuracy | Hurt |
| constrain embeddings to the unit sphere (cosine-style) instead of unconstrained (ablation) | triplet-loss embeddings · synthetic sphere points | triplet accuracy | No clear effect |
Closest papers by meaning
- Embedding Dimension of Contrastive Learning and $k$-Nearest Neighbors (NeurIPS 2024 · similarity 0.87)
- Beyond Worst-Case Dimensionality Reduction for Sparse Vectors (ICLR 2025 · similarity 0.85)
- Polynomial Width is Sufficient for Set Representation with High-dimensional Features (ICLR 2024 · similarity 0.83)
- Trading off Consistency and Dimensionality of Convex Surrogates for Multiclass Classification (NeurIPS 2024 · similarity 0.83)
- Towards Generalization beyond Pointwise Learning: A Unified Information-theoretic Perspective (ICML 2024 · similarity 0.83)
- Optimal Sample Complexity of Contrastive Learning (ICLR 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
