Papers Related to My PhD Thesis-Example Weighting in Classification and Distance Metric Learning
Distance Metric Learning: Learning to Retrieve
- Active Ordinal Querying for Tuplewise Similarity Learning
- Representation Learning with Contrastive Predictive Coding
Learning to Classify
Others
- A general and Adaptive Robust Loss Function-CVPR 2019 Best Paper Finalist
- Unsupervised Embedding Learning via Invariant and Spreading Instance Feature-CVPR 2019
- Regularising Deep Neural Networks by Noise: Its Interpretation and Optimisation-NeurIPS 2017
- Toward Robustness against Label Noise in Training Deep Discriminative Neural Networks-NeurIPS 2017
- Selfie: Refurbishing Unclean Samples for Robust Deep Learning-ICML 2019
- Unsupervised Label Noise Modeling and Loss Correction-ICML 2019
- Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noise-NeurIPS 2018
Example Weighting
- Weighted Machine Learning-Mahdi Hashemi∗, Hassan A. Karimi
- Not All Samples Are Created Equal: Deep Learning with Importance Sampling-Angelos Katharopoulos, Franc¸ois Fleuret, ICML 2018
- computing the importance score for the whole dataset is still prohibitive and would render the method unsuitable for online learning.
- In order to solve the problem of computing the importance for the whole dataset, we pre-sample a large batch of data points, compute the sampling distribution for that batch and re-sample a smaller batch with replacement.
