Research map / Learning theory and algorithms / Sample complexity
On Finite-Sample Identifiability of Contrastive Learning-Based Nonlinear Independent Component Analysis
Qi Lyu, Xiao Fu
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 |
|---|---|---|---|
| Increase network width R from 4 to 64 (ablation) | GCL-based nICA · Synthetic TCL | mutual information | Improved |
| Increase network width beyond 64 (ablation) | GCL-based nICA · Synthetic TCL | mutual information | Hurt |
| Use width R≥128 with N=5000, or R≥256 with N=10000 (ablation) | GCL-based nICA · Synthetic MVCL | mutual information | Hurt |
| Increase network width R (ablation) | GCL-based nICA features · EEG eye | classification error | Mixed |
Closest papers by meaning
- On the Identifiability of Nonlinear ICA: Sparsity and Beyond (NeurIPS 2022 · similarity 0.92)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.91)
- Robustness of Nonlinear Representation Learning (ICML 2024 · similarity 0.90)
- Beyond ICA: Identifiability by Symmetry Breaking (NeurIPS 2026 · similarity 0.87)
- Multi-View Causal Representation Learning with Partial Observability (ICLR 2024 · similarity 0.86)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (NeurIPS 2024 · similarity 0.85)
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.
