pith. sign in

arxiv: 1710.03839 · v1 · pith:OR5S5VWXnew · submitted 2017-10-10 · 💻 cs.LG · cs.IT· math.IT

Disentangled Representations via Synergy Minimization

classification 💻 cs.LG cs.ITmath.IT
keywords representationssynergydisentanglefactorscharactersmakeminimallypredictions
0
0 comments X
read the original abstract

Scientists often seek simplified representations of complex systems to facilitate prediction and understanding. If the factors comprising a representation allow us to make accurate predictions about our system, but obscuring any subset of the factors destroys our ability to make predictions, we say that the representation exhibits informational synergy. We argue that synergy is an undesirable feature in learned representations and that explicitly minimizing synergy can help disentangle the true factors of variation underlying data. We explore different ways of quantifying synergy, deriving new closed-form expressions in some cases, and then show how to modify learning to produce representations that are minimally synergistic. We introduce a benchmark task to disentangle separate characters from images of words. We demonstrate that Minimally Synergistic (MinSyn) representations correctly disentangle characters while methods relying on statistical independence fail.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Now You (Still) See Me: Detecting Evasive Steganographic Payloads in LLMs

    cs.CR 2026-06 unverdicted novelty 6.0

    Adversarial fine-tuning evades activation-based steganography detection in five LLMs while preserving secret recovery, but a recontextualization dataset restores both ridge and MLP probe detectability.