Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Authors: Kaidi Cao, Colin Wei, Adrien Gaidon, Nikos Arechiga, Tengyu Ma
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We test our methods on several benchmark vision tasks including the real-world imbalanced dataset i Naturalist 2018. Our experiments show that either of these methods alone can already improve over existing techniques and their combination achieves even better performance gains. (Abstract) |
| Researcher Affiliation | Collaboration | Kaidi Cao Stanford University kaidicao@stanford.edu Colin Wei Stanford University colinwei@stanford.edu Adrien Gaidon Toyota Research Institute adrien.gaidon@tri.global Nikos Arechiga Toyota Research Institute nikos.arechiga@tri.global Tengyu Ma Stanford University tengyuma@stanford.edu |
| Pseudocode | Yes | Algorithm 1 Deferred Re-balancing Optimization with LDAM Loss (Section 3.3) |
| Open Source Code | Yes | 1Code available at https://github.com/kaidic/LDAM-DRW. (Footnote on page 1) |
| Open Datasets | Yes | We evaluate our proposed algorithm on artificially created versions of IMDB review [41], CIFAR-10, CIFAR-100 [29] and Tiny Image Net [45, 1] with controllable degrees of data imbalance, as well as a real-world large-scale imbalanced dataset, i Naturalist 2018 [52]. (Section 4) |
| Dataset Splits | Yes | To create their imbalanced version, we reduce the number of training examples per class and keep the validation set unchanged. (Section 4.2) We adopt the official training and validation splits for our experiments. (Section 4.3) |
| Hardware Specification | No | The paper does not specify the particular hardware components (e.g., GPU models, CPU types, memory) used for running the experiments. |
| Software Dependencies | No | The paper states 'Our core algorithm is developed using Py Torch [44]' but does not provide a specific version number for PyTorch or any other software dependencies. |
| Experiment Setup | No | The paper mentions using 'Adam optimizer' and 'Res Net-32' or 'Res Net-50' as backbone networks, and learning rate decay 'at epoch 160'. However, it does not provide specific hyperparameter values such as initial learning rate, batch size, or total epochs for all experiments in the main text. |