Deep Learning
Non-normal Recurrent Neural Network (nnRNN): learning long time dependencies while improving expressivity with transient dynamics
Giancarlo Kerg, Kyle Goyette, Maximilian Puelma Touzel, Gauthier Gidel, Eugene Vorontsov, Yoshua Bengio, Guillaume Lajoie
Neural Information Processing Systems http://nips.cc/
Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samples
However, current methods for evaluating such models remain incomplete: standard likelihood-based metrics do not always apply and rarely correlate with perceptual fidelity, while sample-based metrics, such as FID, are insensitive to overfitting, i.e., inability to generalize beyond the training set.
To Trust Or Not To Trust A Classifier
Heinrich Jiang, Been Kim, Melody Guan, Maya Gupta
Neural Information Processing Systems http://nips.cc/
PICProp: Physics-Informed Confidence Propagation for Uncertainty Quantification
Standard approaches for uncertainty quantification in deep learning and physics-informed learning have persistent limitations.
Object-Oriented Dynamics Predictor
Guangxiang Zhu, Zhiao Huang, Chongjie Zhang
Neural Information Processing Systems http://nips.cc/
Input-Output Equivalence of Unitary and Contractive RNNs
Melikasadat Emami, Mojtaba Sahraee Ardakan, Sundeep Rangan, Alyson K. Fletcher
Neural Information Processing Systems http://nips.cc/