Neural Communication Systems with Bandwidth-limited Channel

Abstract

Reliably transmitting messages despite information loss due to a noisy channel is a core problem of information theory. One of the most important aspects of real world communication, e.g. via wifi, is that it may happen at varying levels of information transfer. The bandwidth-limited channel models this phenomenon. In this study we consider learning coding with the bandwidth-limited channel (BWLC). Recently, neural communication models such as variational autoencoders have been studied for the task of source compression. We build upon this work by studying neural communication systems with the BWLC. Specifically,we find three modelling choices that are relevant under expected information loss. First, instead of separating the sub-tasks of compression (source coding) and error correction (channel coding), we propose to model both jointly. Framing the problem as a variational learning problem, we conclude that joint systems outperform their separate counterparts when coding is performed by flexible learnable function approximators such as neural networks. To facilitate learning, we introduce a differentiable and computationally efficient version of the bandwidth-limited channel. Second, we propose a design to model missing information with a prior, and incorporate this into the channel model. Finally, sampling from the joint model is improved by introducing auxiliary latent variables in the decoder. Experimental results justify the validity of our design decisions through improved distortion and FID scores.

References in corpus (19)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Neural Discrete Representation Learning
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Deep Joint Source-Channel Coding for Wireless Image Transmission
• Variational image compression with a scale hyperprior
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Deep Learning-Based Communication Over the Air
• Image and Video Compression with Neural Networks: A Review
• Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations
• Joint Autoregressive and Hierarchical Priors for Learned Image Compression
• Density Modeling of Images using a Generalized Normalization Transformation
• Lossy Image Compression with Compressive Autoencoders
• Variational inference for Monte Carlo objectives
• Towards Image Understanding from Deep Compression without Decoding
• Practical Lossless Compression with Latent Variables using Bits Back Coding
• Fixing a Broken ELBO
• Neural Joint Source-Channel Coding
• Hierarchical Autoregressive Image Models with Auxiliary Decoders
• Associative Compression Networks for Representation Learning