Multi-resolution sources exacerbate this challenge due to complex correlations and sample complexity that scales in the length of the sequence.
All scene graph models to date are limited to training on a small set of visual relationships that have thousands of training labels each.
Ranked #1 on Scene Graph Detection on VRD
Using modern deep learning models to make predictions on time series data from wearable sensors generally requires large amounts of labeled data.
Labeling training data is a key bottleneck in the modern machine learning pipeline.
Training accurate classifiers requires many labels, but each label provides only limited information (one bit for binary classification).
Obtaining enough labeled data to robustly train complex discriminative models is a major bottleneck in the machine learning pipeline.
Prior work has explored learning accuracies for these sources even without ground truth labels, but they assume that a single accuracy parameter is sufficient to model the behavior of these sources over the entire training set.