Factor-analytic inverse regression for high-dimension, small-sample dimensionality reduction
Authors: Aditi Jha, Michael J. Morais, Jonathan W Pillow
ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To test the effectiveness of CFAD as compared to other inverse regression SDR methods, we simulate observations from a CFAD model...We demonstrate the effectiveness of CFAD with an application to functional magnetic resonance imaging (f MRI) measurements during visual object recognition and working memory tasks, where it outperforms existing SDR and a variety of other dimensionality-reduction methods. |
| Researcher Affiliation | Academia | 1Princeton Neuroscience Institute, Princeton University, NJ, USA 2Department of Electrical and Computer Engineering, Princeton University, NJ, USA 3Department of Psychology, Princeton University, NJ, USA. |
| Pseudocode | No | No pseudocode or algorithm blocks were found in the paper. |
| Open Source Code | Yes | Code available at: https://github.com/97aditi/CFAD.git |
| Open Datasets | Yes | We first used CFAD to classify functional magnetic resonance imaging (f MRI) activity recorded during a visual object recognition task (Haxby et al., 2001)...Further details about the dataset can be found in Barch et al., 2013; WU Minn Consortium Human Connectome Project, 2017. |
| Dataset Splits | Yes | We perform a 5-fold cross-validation...With each method, we perform a 3-fold cross validation analysis... |
| Hardware Specification | No | No specific hardware details (e.g., GPU/CPU models, memory specifications) used for running the experiments were provided in the paper. Only general descriptions of tasks and methods were given. |
| Software Dependencies | No | The paper mentions the 'nilearn package (Abraham et al., 2014)' and 'pymanopt library (Townsend et al., 2016)', but does not provide specific version numbers for these or other software dependencies. |
| Experiment Setup | Yes | For all methods, except RRR, we vary d in increments of 10 in (0, p]...We initialize s CFAD and CFAD with SIR, using the initialization scheme discussed in Section. 3.1 and choose the smoothness hyper-parameter λ [10 3, 104] by nested cross-validation within each of the 5 folds. |