6 papers
Singular Relative Entropy Coding with Bits-Back Rejection Sampling
Gergely Flamich, Spencer Hill
A relative entropy code for a source is a stochastic code that encodes random samples from a prescribed using as few bits as possible. A generalisation…
Data Compression with Relative Entropy Coding
Gergely Flamich
Over the last few years, machine learning unlocked previously infeasible features for compression, such as providing guarantees for users' privacy or tailoring compression to speci…
You Cannot Feed Two Birds with One Score: the Accuracy-Naturalness Tradeoff in Translation
Gergely Flamich, David Vilar, Jan-Thorsten Peter +1
The goal of translation, be it by human or by machine, is, given some text in a source language, to produce text in a target language that simultaneously 1) preserves the meaning o…
The Redundancy of Non-Singular Channel Simulation
Gergely Flamich, Sharang M. Sriramu, Aaron B. Wagner
Channel simulation is an alternative to quantization and entropy coding for performing lossy source coding. Recently, channel simulation has gained significant traction in both the…
Getting Free Bits Back from Rotational Symmetries in LLMs
Jiajun He, Gergely Flamich, José Miguel Hernández-Lobato
Current methods for compressing neural network weights, such as decomposition, pruning, quantization, and channel simulation, often overlook the inherent symmetries within these ne…
Compressing Images by Encoding Their Latent Representations with Relative Entropy Coding
Gergely Flamich, Marton Havasi, José Miguel Hernández-Lobato
Variational Autoencoders (VAEs) have seen widespread use in learned image compression. They are used to learn expressive latent representations on which downstream compression meth…