Hierarchical harmonium

Identity Notes & Concept Scope Scoped to Yang and colleagues' shared hierarchical harmonium for semantic video. The label layer couples to the same hidden topics as text and images, unlike FoH's independently parameterized class models. Although its diagram has three tiers, the graph is bipartite between the latent variables and the combined text/image/label layer. The source uses one-hot labeled training shots and a single-label decision rule, while its Bernoulli label conditionals define a joint over the full binary label cube; the conclusion explicitly permits multiple active categories. This is not an ontology class hierarchy or an asserted one-hot generative model. No normalized identity matched four candidates against all 442 pinned AIO IDs, labels and scoped synonyms or 444 preceding native IDs, names and aliases. Full-checkout searches included ignored and hidden files in both checkouts; five matching lines per checkout came from two unique prior discovery-ledger contents, and all matching excerpts were read. All 205 semantic label/synonym hits were byte-verified against prior readings. Twenty-nine annotated AIO nodes and outgoing edges comprise 28 unchanged prior readings and one newly read multimodal-deep-learning node. Thirty-nine contextual YAMLs comprise 34 unchanged prior readings and five fully read current records. Three candidates are curated; the unbounded-count DWH specialization remains deferred because the inspected joint is not normalizable with nonzero count couplings. Absence is bounded to this pinned snapshot and maintained corpus. Unlike the maintained conditional-RBM record, which conditions on an externally supplied input and curates binary hidden units, HH specifies a joint over text, images and labels with Gaussian hidden variables.