Research map / Probabilistic and scientific modeling / Causal discovery
Independent mechanism analysis, a new concept?
Luigi Gresele, Julius Von Kügelgen, Vincent Stimper, Bernhard Schölkopf, Michel Besserve
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 |
|---|---|---|---|
| add C_IMA regularisation (lambda 0.5 or 1.0) to maximum-likelihood training of a residual flow (ablation) | residual flow nonlinear ICA · synthetic Mobius mixing, n=5 and n=7 | nonlinear Amari distance (lower is better) and MCC | Improved |
| increase lambda from 0.5 to 1.0 (ablation) | C_IMA-regularised MLE · synthetic Mobius mixing, n=5 and n=7 | nonlinear Amari distance and MCC | No clear effect |
| vary lambda in {0,0.5,1} (ablation) | C_IMA-regularised MLE · synthetic Mobius mixing, n=5 and n=7 | KL divergence to true likelihood | No clear effect |
| increase number of mixing-MLP layers L from 2 to 4 | true mixing function · random invertible MLP mixing, n=5 | C_IMA of true mixing | Hurt |
| increase dimension n from 2 to 10 | learnt Darmois solutions · Mobius mixing, n in {2,3,5,10} | C_IMA | Mixed |
Closest papers by meaning
- Identifiable Exchangeable Mechanisms for Causal Structure and Representation Learning (ICLR 2025 · similarity 0.89)
- Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability (NeurIPS 2025 · similarity 0.88)
- Learning Linear Causal Representations from General Environments: Identifiability and Intrinsic Ambiguity (NeurIPS 2024 · similarity 0.88)
- Robustness of Nonlinear Representation Learning (ICML 2024 · similarity 0.88)
- Synergy Between Sufficient Changes and Sparse Mixing Procedure for Disentangled Representation Learning (ICLR 2025 · similarity 0.87)
- Multi-View Causal Representation Learning with Partial Observability (ICLR 2024 · similarity 0.87)
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.
