Notice: 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. Full accuracy metrics and methodology are described in [1].
Meta-Learning with Memory-Augmented Neural Networks
Authors: Adam Santoro, Sergey Bartunov, Matthew Botvinick, Daan Wierstra, Timothy Lillicrap
ICML 2016 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the ability of a memory-augmented neural network to rapidly assimilate new data, and leverage this data to make accurate predictions after only a few samples. We also introduce a new method for accessing an external memory that focuses on memory content, unlike previous methods that additionally use memory locationbased focusing mechanisms. |
| Researcher Affiliation | Collaboration | Adam Santoro EMAIL Google Deep Mind Sergey Bartunov EMAIL Google Deep Mind, National Research University Higher School of Economics (HSE) Matthew Botvinick EMAIL Daan Wierstra EMAIL Timothy Lillicrap EMAIL Google Deep Mind |
| Pseudocode | No | The paper does not contain any structured pseudocode or clearly labeled algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement about making source code available or include a link to a code repository. |
| Open Datasets | Yes | The Omniglot dataset consists of over 1600 separate classes with only a few examples per class, aptly lending to it being called the transpose of MNIST (Lake et al., 2015). |
| Dataset Splits | No | The paper mentions training on 1200 original classes and using 423 classes for test experiments but does not explicitly describe a validation set split, specific percentages for splits, or cross-validation setup. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running the experiments. |
| Software Dependencies | No | The paper mentions architectures like Neural Turing Machines (NTMs) and LSTMs, but does not provide specific software dependencies or version numbers (e.g., Python, PyTorch, TensorFlow versions or other libraries with versions). |
| Experiment Setup | Yes | After training on 100,000 episodes with five randomly chosen classes with randomly chosen labels, the network was given a series of test episodes. |