On the Generalization and Approximation Capacities of Neural Controlled Differential Equations
Authors: Linus Bleistein, Agathe Guilloux
ICLR 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our theoretical results are validated through a series of experiments. Our theoretical results are illustrated by experiments on synthetic data. |
| Researcher Affiliation | Academia | Linus Bleistein Inria Paris, UEVE Agathe Guilloux Inria Paris |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Our code is available at this link. |
| Open Datasets | No | The paper describes generating synthetic data using 'Python package stochastic' and 'f BM with Hurst parameter H'. It does not provide a specific link, DOI, repository name, or formal citation for a pre-existing, publicly available dataset. |
| Dataset Splits | No | The paper mentions training sample sizes and test sample sizes but does not specify a validation split or its size: 'The size of the training sample is set to n = 100.' and 'the expected generalization error is computed on 50 test samples.' |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, memory, or cloud instance types used for experiments. |
| Software Dependencies | No | The paper mentions 'Python package stochastic' and 'Pytorch s default initialization' but does not provide specific version numbers for any software dependencies. |
| Experiment Setup | Yes | The model is initialized with Pytorch s default initialization. In the first and second figures (starting from the left), the model is trained for 2000 iterations with Adam. We use the default values for α, β and a learning rate of 5 × 10−3. The size of the training sample is set to n = 100. We train a shallow NCDE classifier with p = 3 on n = 100 time series sampled at 100 equidistant time points in [0, 1] for 100 iterations with Binary Cross Entropy (BCE) loss. We use Adam with default settings and a learning rate of 5 × 10−2. |