Embedding Probabilities in Predication Space with Hermitian Holographic Reduced Representations
نویسندگان
چکیده
Predication-based Semantic Indexing (PSI) is an approach to generating high-dimensional vector representations of concept-relation-concept triplets. In this paper, we develop a variant of PSI that accommodates estimation of the probability of encountering a particular predication (such as fluoxetine TREATS major depressive disorder) in a collection of predications concerning a concept of interest (such as major depressive disorder). PSI leverages reversible vector transformations provided by representational approaches known as Vector Symbolic Architectures (VSA). To embed probabilities we develop a novel VSA variant, Hermitian Holographic Reduced Representations, with improvements in predictive modeling experiments. The probabilistic interpretation this facilitates reveals previously unrecognized connections between PSI and quantum theory perhaps most notably that PSI’s estimation of relatedness across multiple reasoning pathways corresponds to the estimation of the probability of traversing indistinguishable pathways in accordance with the rules of quantum probability.
منابع مشابه
Maximal Surface Group Representations in Isometry Groups of Classical Hermitian Symmetric Spaces
Higgs bundles and non-abelian Hodge theory provide holomorphic methods with which to study the moduli spaces of surface group representations in a reductive Lie group G. In this paper we survey the case in which G is the isometry group of a classical Hermitian symmetric space of non-compact type. Using Morse theory on the moduli spaces of Higgs bundles, we compute the number of connected compon...
متن کاملOn a Class of Hyperplanes of the Symplectic and Hermitian Dual Polar Spaces
Let ∆ be a symplectic dual polar space DW (2n−1, K) or a Hermitian dual polar space DH(2n − 1, K, θ), n ≥ 2. We define a class of hyperplanes of ∆ arising from its Grassmann-embedding and discuss several properties of these hyperplanes. The construction of these hyperplanes allows us to prove that there exists an ovoid of the Hermitian dual polar space DH(2n−1, K, θ) arising from its Grassmann-...
متن کاملExperiments on Preserving Pieces of Information in a Given Order in Holographic Reduced Representations and the Continuous Geometric Algebra Model
Geometric Analogues of Holographic Reduced Representations (GAc, which is the continuous version of the previously developed discrete GA model) employ role-filler binding based on geometric products. Atomic objects are real-valued vectors in n-dimensional Euclidean space and complex statements belong to a hierarchy of multivectors. The property of GAc and HRR studied here is the ability to stor...
متن کاملHolographic Embeddings of Knowledge Graphs
Learning embeddings of entities and relations is an efficient and versatile method to perform machine learning on relational data such as knowledge graphs. In this work, we propose holographic embeddings (HOLE) to learn compositional vector space representations of entire knowledge graphs. The proposed method is related to holographic models of associative memory in that it employs circular cor...
متن کاملA Note on Holographic Renormalization of Probe D-Branes
A great deal of progress has been recently made in the study of holography for non-conformal branes. Considering the near-horizon limit of backgrounds generated by such branes, we discuss the holographic renormalization of probe D-branes in these geometries. More specifically, we discuss in some detail systems with a codimension-one defect. For this class of systems, the mode which describes th...
متن کامل