Lossless compression with state space models using bits back coding

James Townsend, Iain Murray

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We generalize the ‘bits back with ANS’ method to time-series models with a latent Markov structure. This family of models includes hidden Markov models (HMMs), linear Gaussian state space models (LGSSMs) and many more. We provide experimental evidence that our method is effective for small scale models, and discuss its applicability to larger scale settings such as video compression.
Original languageEnglish
Title of host publicationNeural Compression Workshop at ICLR 2021
Number of pages5
Publication statusAccepted/In press - 27 Mar 2021
EventNeural Compression Workshop @ ICLR 2021 - Online
Duration: 7 May 20217 May 2021
https://neuralcompression.github.io/

Workshop

WorkshopNeural Compression Workshop @ ICLR 2021
Period7/05/217/05/21
Internet address

Fingerprint Dive into the research topics of 'Lossless compression with state space models using bits back coding'. Together they form a unique fingerprint.

Cite this