Learning with Feature-Dependent Label Noise: A Progressive Approach
Authors: Yikai Zhang, Songzhu Zheng, Pengxiang Wu, Mayank Goswami, Chao Chen
ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In experiments, our method outperforms SOTA baselines and is robust to various noise types and levels. We evaluate our method on both synthetic and real-world datasets. |
| Researcher Affiliation | Academia | 1Rutgers University, {yz422,pw241}@cs.rutgers.edu 2Stony Brook University, {zheng.songzhu,chao.chen.1}@stonybrook.edu 3City University of New York, mayank.isi@gmail.com |
| Pseudocode | Yes | Algorithm 1 Progressive Label Correction |
| Open Source Code | Yes | Our code is available at https://github.com/pxiangwu/PLC. |
| Open Datasets | Yes | We first conduct synthetic experiments on two public datasets CIFAR-10 and CIFAR-100 (Krizhevsky et al., 2009). To test the effectiveness of the proposed method under realworld label noise, we conduct experiments on the Clothing1M dataset (Xiao et al., 2015). Apart from Clothing1M, we also test our method on another smaller dataset, Food-101N (Lee et al., 2018). Finally, we test our method on a recently proposed real-world dataset, ANIMAL-10N (Song et al., 2019). |
| Dataset Splits | Yes | This dataset provides 50k, 14k and 10k manually verified clean data for training, validation and testing, respectively. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory specifications) used for running the experiments. |
| Software Dependencies | No | The paper mentions using a 'Res Net-50' and 'SGD optimizer' but does not specify software versions for programming languages, frameworks, or libraries (e.g., Python version, PyTorch/TensorFlow version, CUDA version). |
| Experiment Setup | Yes | During training, we use a batch size of 128 and train the network for 180 epochs to ensure the convergence of all methods. We train the network with SGD optimizer, with initial learning rate 0.01. We set the batch size 32, learning rate 0.001, and adopt SGD optimizer and We train the network for 30 epochs with SGD optimizer. The batch size is 32 and the initial learning rate is 0.005, which is divided by 10 every 10 epochs. and we train the network for 100 epochs and use an initial learning rate of 0.1, which is divided by 5 at 50% and 75% of the total number of epochs. |