Provable Multi-instance Deep AUC Maximization with Stochastic Pooling
Jul 24, 2023
Speakers
Dixian Zhu
Speaker · 0 followers
Bokun Wang
Speaker · 0 followers
Yaxing Wang
Speaker · 0 followers
Milan Sonka
Speaker · 0 followers
Xiaodong Wu
Speaker · 0 followers
Tianbao Yang
Speaker · 0 followers
About
This paper considers a novel application of deep AUC maximization (DAM) for multi-instance learning (MIL), in which a single class label is assigned to a bag of instances (e.g., multiple 2D slices of a CT scan for a patient). We address a neglected yet non-negligible computational challenge of MIL in the context of DAM, i.e., bag size is too large to be loaded into GPU memory for back-propagation, which is required by the standard pooling methods of MIL. To tackle this challenge, we propose variance-reduced stochastic pooling methods in the spirit of stochastic optimization by formulating the loss function over the pooled prediction as a multi-level compositional function. By synthesizing techniques from stochastic compositional optimization and non-convex min-max optimization, we propose a unified and provable muli-instance DAM (MIDAM) algorithm with stochastic smoothed-max pooling or stochastic attention-based pooling, which only samples a few instances for each bag to compute a stochastic gradient estimator and to update the model parameter. We establish a similar convergence rate of the proposed MIDAM algorithm as the state-of-the-art DAM algorithms. Our extensive experiments on conventional MIL datasets and medical datasets demonstrate the superiority of our MIDAM algorithm.This paper considers a novel application of deep AUC maximization (DAM) for multi-instance learning (MIL), in which a single class label is assigned to a bag of instances (e.g., multiple 2D slices of a CT scan for a patient). We address a neglected yet non-negligible computational challenge of MIL in the context of DAM, i.e., bag size is too large to be loaded into GPU memory for back-propagation, which is required by the standard pooling methods of MIL. To tackle this challenge, we propose vari…
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
Revolutionizing Multimedia Applications with Neural Codecs and Foundation Models
Watch later
Global Selection of Contrastive Batches via Optimization on Sample Permutations
Watch later
Differentiable Rendering and Beyond
Watch later
Latent Masking for Multimodal Self-supervised Learning in Health Timeseries
Watch later
Make-A-Video 3D (MAV3D): Text-To-4D Dynamic Scene Generation
Watch later
Uriel Singer, …
