Reproduce “An information-geometric formulation of distributed computation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Alberto Liardi, George Blackburne, Pedro A.M. Mediano
The authors introduce ΦID_IG, a framework that decomposes input–output information into source–target atoms using geometric synergies derived from conditional-independence projections. It distinguishes information carried individually from information requiring joint contributions, tracing these modes across a system. In an S&P 500 analysis, S→S modularity aligned with sectors more than null networks, unlike total information; this is an association, not causation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.