Contrastive Learning Inverts the Data Generating Process

Authors: Roland S. Zimmermann, Yash Sharma, Steffen Schneider, Matthias Bethge, Wieland Brendel

ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically verify our predictions when the assumed theoretical conditions are fulfilled. In addition, we show successful inversion of the data generating process even if these theoretical assumptions are partially violated. Tables 1 and 2 show results evaluating identifiability up to affine transformations and generalized permutations, respectively.
Researcher Affiliation Academia 1University of T ubingen, T ubingen, Germany 2IMPRS for Intelligent Systems, T ubingen, Germany 3EPFL, Geneva, Switzerland.
Pseudocode No No structured pseudocode or algorithm blocks were found in the paper.
Open Source Code Yes 1Online version and code: brendel-group.github.io/cl-ica/
Open Datasets Yes 3DIdent is available at zenodo.org/record/4502485.
Dataset Splits No The paper mentions a test set for 3DIdent but does not explicitly provide details for training/validation/test splits (e.g., percentages or sample counts for each split).
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions software like Blender and FAISS by name and citation, but does not provide specific version numbers for these or other software dependencies used in the experiments.
Experiment Setup Yes For further details, see Appx. A.3. In our experiments, we use the same training hyperparameters (for details see Appx. A.3) and (encoder) architecture as Klindt et al. (2021).