End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information

12/11/2019

∙

GMI-based end-to-end learning is shown to be highly nonconvex. We apply gradient descent initialized with Gray-labeled APSK constellations directly to the constellation coordinates. State-of-the-art constellations in 2D and 4D are found providing reach increases up to 26% w.r.t. to QAM.

READ FULL TEXT

End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information

End-to-End Learning of Joint Geometric and Probabilistic Constellation Shaping

Geometric Constellation Shaping for Fiber-Optic Channels via End-to-End Learning

Simplified End-to-End MMI Training and Voting for ASR

Learning Unbiased Representations via Mutual Information Backpropagation

Generalized Bregman Divergence and Gradient of Mutual Information for Vector Poisson Channels

Maximizing Mutual Information for Tacotron

End-to-End Efficient Representation Learning via Cascading Combinatorial Optimization

End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information

Related Research

End-to-End Learning of Joint Geometric and Probabilistic Constellation Shaping

Geometric Constellation Shaping for Fiber-Optic Channels via End-to-End Learning

Simplified End-to-End MMI Training and Voting for ASR

Learning Unbiased Representations via Mutual Information Backpropagation

Generalized Bregman Divergence and Gradient of Mutual Information for Vector Poisson Channels

Maximizing Mutual Information for Tacotron

End-to-End Efficient Representation Learning via Cascading Combinatorial Optimization