Phase Transitions in Rate Distortion Theory and Deep Learning.
Philipp GrohsAndreas KlotzFelix VoigtländerPublished in: CoRR (2020)
Keyphrases
- deep learning
- phase transition
- rate distortion theory
- information theory
- rate distortion
- spatially adaptive
- unsupervised learning
- constraint satisfaction
- np complete
- mental models
- machine learning
- weakly supervised
- bit rate
- information theoretic
- random constraint satisfaction problems
- video coding
- denoising
- motion estimation
- color images
- lower bound
- compressed images
- image data
- learning algorithm