5 papers · 1 filter
Matching High-Dimensional Geometric Quantiles for Test-Time Adaptation of Transformers and Convolutional Networks Alike
Sravan Danda, Aditya Challa, Shlok Mehendale +1
Test-time adaptation (TTA) refers to adapting a classifier for the test data when the probability distribution of the test data slightly differs from that of the training data of t…
A Radon-Nikodým Perspective on Anomaly Detection: Theory and Implications
Shlok Mehendale, Aditya Challa, Rahul Yedida +3
Which principle underpins the design of an effective anomaly detection loss function? The answer lies in the concept of Radon-Nikodým theorem, a fundamental concept in measure the…
Quantile Activation: Correcting a Failure Mode of ML Models
Aditya Challa, Sravan Danda, Laurent Najman +1
Standard ML models fail to infer the context distribution and suitably adapt. For instance, the learning fails when the underlying distribution is actually a mixture of distributio…
A Granger-Causal Perspective on Gradient Descent with Application to Pruning
Aditya Shah, Aditya Challa, Sravan Danda +2
Stochastic Gradient Descent (SGD) is the main approach to optimizing neural networks. Several generalization properties of deep networks, such as convergence to a flatter minima, a…
QuantProb: Generalizing Probabilities along with Predictions for a Pre-trained Classifier
Aditya Challa, Snehanshu Saha, Soma Dhavala
Quantification of Uncertainty in predictions is a challenging problem. In the classification settings, although deep learning based models generalize well, class probabilities ofte…