{"slug": "physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for", "title": "Physics Closure Matters for Machine Olfaction: A Maxwell--Stefan Graph Solver for Identifiable Dynamic Gas Unmixing", "summary": "Researchers propose UnMixNet, a physics-closed graph neural solver that embeds Maxwell-Stefan multicomponent transport PDEs, competitive adsorption ODEs, and nonlinear sensor transduction ODEs into end-to-end gas unmixing, addressing physics closure misspecification in machine olfaction. Evaluations on SmellNet show improved single-odor recognition, seen-mixture unmixing, and unseen-mixture generalization, while external validation on UCI Dynamic Gas Mixtures confirms inferred concentration processes agree with ground truth set points under dynamic transitions.", "body_md": "arXiv:2607.18544v1 Announce Type: new\nAbstract: Machine olfaction for gas unmixing is an underconstrained inverse problem in which gas compositions must be inferred from low-dimensional, delayed, and entangled sensor responses produced by interacting chemical transport, surface adsorption, and sensor transduction. One of the key obstacles is physics closure misspecification, where a neural network is designed to fit sensor traces rather than infer a physically closed olfactory process. In this work, we formulate gas unmixing as a multi-physics-constrained inverse problem governed by Maxwell--Stefan multicomponent transport PDEs, competitive adsorption ODEs, and nonlinear sensor transduction ODEs. Directly solving such a high-dimensional coupled system is computationally expensive and often numerically unstable. To this end, we propose UnMixNet, a physics-closed graph neural solver that embeds this multi-physics forward process into end-to-end gas unmixing. UnMixNet discretizes Maxwell--Stefan cross-diffusion on spatial graphs and formulates the multicomponent flux on each edge. This design enables local, differentiable, and flux-conservative inference for multicomponent cross-diffusion. Evaluations on SmellNet show improved single-odor recognition, seen-mixture unmixing, and unseen-mixture generalization. In addition, an external validation on UCI Dynamic Gas Mixtures shows that the inferred concentration process agrees with ground truth concentration set points under dynamic transitions. Process-consistency diagnostics further show that the proposed model learns transferable dynamic physical fingerprints that better satisfies transport, conservation, adsorption, and readout closure.", "url": "https://wpnews.pro/news/physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for", "canonical_source": "https://arxiv.org/abs/2607.18544", "published_at": "2026-07-22 04:00:00+00:00", "updated_at": "2026-07-22 04:13:59.416144+00:00", "lang": "en", "topics": ["machine-learning", "neural-networks", "artificial-intelligence"], "entities": ["UnMixNet", "SmellNet", "UCI Dynamic Gas Mixtures", "Maxwell-Stefan"], "alternates": {"html": "https://wpnews.pro/news/physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for", "markdown": "https://wpnews.pro/news/physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for.md", "text": "https://wpnews.pro/news/physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for.txt", "jsonld": "https://wpnews.pro/news/physics-closure-matters-for-machine-olfaction-a-maxwell-stefan-graph-solver-for.jsonld"}}