Multi-View Causal Representation Learning with Partial Observability
Authors: Dingling Yao, Danru Xu, Sebastien Lachapelle, Sara Magliacane, Perouz Taslakian, Georg Martius, Julius von Kügelgen, Francesco Locatello
ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimentally validate our claims on numerical, image, and multimodal data sets. |
| Researcher Affiliation | Collaboration | 1Institute of Science and Technology Austria 2Max Planck Institute for Intelligent Systems, Tübingen 3University of Amsterdam 4University of Tübingen 5University of Cambridge 6MIT-IBM Watson AI Lab 7Samsung SAIT AI Lab 8Mila, Université de Montréal 9Service Now Research |
| Pseudocode | No | The paper does not contain any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code is available at https://github.com/Causal Learning AI/multiview-crl. |
| Open Datasets | Yes | We test on MPI-3D complex (Gondal et al., 2019) and 3DIdent (Zimmermann et al., 2021). ... Multimodal3DIdent (Daunhawer et al., 2023) augments Causal3DIdent (von Kügelgen et al., 2021) with text annotations for each image view... |
| Dataset Splits | No | The paper mentions optimizing hyperparameters over "different validations" but does not specify the exact training, validation, and test dataset splits (e.g., percentages or sample counts). |
| Hardware Specification | No | The paper mentions support for using "the Dutch National Supercomputer Snellius" in the acknowledgments, but it does not provide specific hardware details such as GPU/CPU models, processor types, or memory specifications used for running the experiments. |
| Software Dependencies | No | The paper mentions using the Adam optimizer and building upon code from other papers, but it does not provide specific version numbers for software dependencies such as Python, PyTorch, or CUDA. |
| Experiment Setup | Yes | Table 4: Parameters for numerical simulation ( 5.1 and App. D.1). ... Table 5: Parameters for experiments 5.2 and 5.3 and Apps. D.3 and D.4. |