Evidential Conditional Neural Processes
Authors: Deep Shankar Pandey, Qi Yu
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments conducted on both synthetic and real-world data demonstrate the effectiveness of our proposed model in various few-shot settings. Experiments Datasets. For function regression experiments, we consider two synthetic datasets i) sinusoidal function regression (Gondal et al. 2021), and ii) regression on sample functions from a Gaussian process (Garnelo et al. 2018a). For Image completion experiments, we consider three benchmark datasets: MNIST (Deng 2012) Celeb A (Liu et al. 2015), and Cifar10 (Krizhevsky, Hinton et al. 2009). |
| Researcher Affiliation | Academia | Deep Shankar Pandey, Qi Yu Rochester Institute of Technology {dp7972, qi.yu}@rit.edu |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about the release of source code for the described methodology or a link to a code repository. |
| Open Datasets | Yes | For function regression experiments, we consider two synthetic datasets i) sinusoidal function regression (Gondal et al. 2021), and ii) regression on sample functions from a Gaussian process (Garnelo et al. 2018a). For Image completion experiments, we consider three benchmark datasets: MNIST (Deng 2012) Celeb A (Liu et al. 2015), and Cifar10 (Krizhevsky, Hinton et al. 2009). |
| Dataset Splits | No | The paper describes meta-training and meta-testing phases, and defines context and target sets within each task. It specifies that models are trained on meta-training tasks and evaluated on meta-testing tasks, which comprise context and target sets. However, it does not explicitly mention or specify a distinct 'validation' dataset split for hyperparameter tuning, beyond the internal task structure. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running the experiments, such as GPU/CPU models or memory specifications. |
| Software Dependencies | No | The paper does not provide specific software dependency details, such as library names with version numbers, needed to replicate the experiment. |
| Experiment Setup | Yes | the function regression models are trained for 30,000 meta-iterations using a batch of 8 tasks and evaluated on 2,000 test tasks. Image completion task is created by randomly selecting a subset of the set points (input-output pairs) from an image. Specifically, each position in the image grid is the input and the pixel value (e.g., the RGB value) is the output. We randomly select 50 points to make the context set, use the remaining points in the image to make the target set, and train models for 50 epochs using a batch of 8 tasks, and evaluate the model on the test set. Both models are trained for 20,000 iterations using training tasks with data in range [ 5, 5]. |