Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
Learning Invariances using the Marginal Likelihood
Authors: Mark van der Wilk, Matthias Bauer, ST John, James Hensman
NeurIPS 2018 | Venue PDF | LLM Run Details | Input Tokens: 17,571 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,254 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 5 Experiments We demonstrate our approach on a series of experiments on variants of the MNIST datasets. While MNIST has been accurately solved by other methods, we intend to show that a model like an RBF GP (Radial Basis Function or squared exponential kernel), for which MNIST is challenging, can be significantly improved by learning the correct invariances. For binary classification tasks, we will use the Pòlya-Gamma approximation for the logistic likelihood, while for multi-class classification, we are currently forced to use the Gaussian likelihood. |
| Researcher Affiliation | Collaboration | Mark van der Wilk PROWLER.io Cambridge, UK EMAIL Matthias Bauer MPI for Intelligent Systems University of Cambridge EMAIL ST John PROWLER.io Cambridge, UK EMAIL James Hensman PROWLER.io Cambridge, UK EMAIL |
| Pseudocode | No | The paper describes procedures and mathematical formulations but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions 'Affine transform implementation from github.com/kevinzakka/spatial-transformer-network.' in a footnote, but this is a third-party tool and there is no explicit statement about releasing the authors' own source code for their methodology. |
| Open Datasets | Yes | We demonstrate our approach on a series of experiments on variants of the MNIST datasets. |
| Dataset Splits | No | Crucially, we do not require a validation set, but can use the log marginal likelihood of the training data to monitor performance. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware used for running the experiments. |
| Software Dependencies | No | The paper mentions the use of Gaussian process models and a Pòlya-Gamma approximation, and references an affine transform implementation, but does not provide specific version numbers for any software dependencies or libraries. |
| Experiment Setup | No | While the paper describes the types of transformations and objectives used in experiments, it does not provide specific hyperparameters such as learning rates, batch sizes, number of epochs, or optimizer settings. |