Telefonaktiebolaget Lm Ericsson (publ)

University of Pennsylvania

Hamed Hassani

Dennis Hui

Ivana Maric

Songnam Hong

George J Pappas

Marco Mondelli

Alexander Robey

Methods, systems, and computer readable media for model-based robust deep learning. In some examples, a method includes obtaining a model of natural variation for a machine learning task. The model of natural variation includes a mapping that specifies how an input datum can be naturally varied by a nuisance parameter. The method includes training, using the model of natural variation and training data for the machine learning task, a neural network to complete the machine learning task such that the neural network is robust to natural variation specified by the model of natural variation.

Model-based robust deep learning

According to certain embodiments, a transmit node in a wireless communications system includes a first universal rate-compatible polar encoder and a transmitter. The first universal rate-compatible polar encoder is configured for a family of two or more types of channels and encodes a plurality of information bits to provide a plurality of coded bits. The transmitter transmits the plurality of coded bits to a receive node.

Systems and methods for rate-compatible polar codes for general channels

Growing community of inventors

Philadelphia, PA, United States of America

Hamed Hassani