Stochastic Optimization of Areas Under Precision-Recall Curves with Provable Convergence
Dec 6, 2021
Speakers
Youzhi Luo
Speaker · 0 followers
Shuiwang Ji
Speaker · 0 followers
Tianbao Yang
Speaker · 0 followers
About
Areas under ROC (AUROC) and precision-recall curves (AUPRC) are common metrics for evaluating classification performance for imbalanced problems. Compared with AUROC, AUPRC is a more appropriate metric for highly imbalanced datasets. While stochastic optimization of AUROC has been studied extensively, principled stochastic optimization of AUPRC has been rarely explored. In this work, we propose a principled technical method to optimize AUPRC for deep learning. Our approach is based on maximizing the averaged precision (AP), which is an unbiased point estimator of AUPRC. We cast the objective into a sum of dependent compositional functions with inner functions dependent on random variables of the outer level. We propose efficient adaptive and non-adaptive stochastic algorithms with provable convergence guarantee under mild conditions by leveraging recent advances in stochastic compositional optimization. Extensive experimental results on image and graph datasets demonstrate that our proposed method outperforms prior methods on imbalanced problems in terms of AUPRC. To the best of our knowledge, our work represents the first attempt to optimize AUPRC with provable convergence.Areas under ROC (AUROC) and precision-recall curves (AUPRC) are common metrics for evaluating classification performance for imbalanced problems. Compared with AUROC, AUPRC is a more appropriate metric for highly imbalanced datasets. While stochastic optimization of AUROC has been studied extensively, principled stochastic optimization of AUPRC has been rarely explored. In this work, we propose a principled technical method to optimize AUPRC for deep learning. Our approach is based on maximizing…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Identification of the Generalized Condorcet Winner in Multi-dueling Bandits
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
The Sensory Neuron as a Transformer: Permutation-Invariant Neural Networks for Reinforcement Learning
Watch later
Yujin Tang, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Leveraging Distribution Alignment via Stein Path for Cross-Domain Cold-Start Recommendation
Watch later
Weiming Liu, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Multi-party referential communication in complex strategic games
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Exploring Social Posterior Collapse in Variational Autoencoder for Interaction Modeling
Watch later
Chen Tang, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
AFEC: Active Forgetting of Negative Transfer in Continual Learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%