Memory Efficient Online Meta Learning
Authors: Durmus Alp Emre Acar, Ruizhao Zhu, Venkatesh Saligrama
ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | On benchmark datasets we show that our method can outperform prior works even though they allow for perfect recall. |
| Researcher Affiliation | Academia | 1Boston University, Boston, MA. Correspondence to: Durmus Alp Emre Acar <alpacar@bu.edu>, Ruizhao Zhu <rzhu@bu.edu>, Venkatesh Saligrama <srv@bu.edu>. |
| Pseudocode | Yes | Algorithm 1 Memory Efficient Online Meta Learning MOML |
| Open Source Code | No | The paper does not include an unambiguous statement that the authors are releasing the code for the work described in this paper, nor does it provide a direct link to a source-code repository. |
| Open Datasets | Yes | Datasets. We evaluate the performance of our approach on three benchmark datasets: MNIST (Le Cun et al., 1998), CIFAR-100 (Krizhevsky et al., 2009) and mini Image Net (Vinyals et al., 2016). |
| Dataset Splits | No | The paper mentions 'train/test set' for each task (e.g., 'train/test set of each task include transformations' for S-MNIST), but explicitly states, 'Hyperparameter tuning in an online setting poses challenges, since unlike the batch setting, we typically do not have a validation set.' |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running its experiments. |
| Software Dependencies | No | The paper mentions software like 'Py Torch framework (Paszke et al., 2019)' and 'Higher (Grefenstette et al., 2019) library' but does not provide specific version numbers for these dependencies. |
| Experiment Setup | No | The paper describes model architectures and states that 'Hyperparameter tuning... leverage the Hedge algorithm... We refer to Appendix A.1 for details of our setup.' However, it does not explicitly provide concrete hyperparameter values (e.g., learning rate, batch size, number of epochs) or specific system-level training settings in the main text. |