TENSKEL: A Combinatorial Observable Tensor for Structured Measurement and Reconstruction

• URL: http://arxiv.org/abs/2608.21999v1
• Date: Sat, 22 Aug 2026 15:10:18 GMT
• Title: TENSKEL: A Combinatorial Observable Tensor for Structured Measurement and Reconstruction
• Abstract summary: TENSKEL is a framework for structured measurement and reconstruction based on tensor representations defined over discrete domains.<n>Rather than introducing a new reconstruction algorithm, this framework provides a mathematical representation of how latent configurations become organized under observation.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many imaging problems seek to reconstruct underlying configurations from partial observable measurements. While reconstruction algorithms operate on these measurements, the observable organization induced by the measurement process is rarely represented explicitly. We introduce TENSKEL, a combinatorial observable framework for structured measurement and reconstruction based on tensor representations defined over discrete domains. Starting from a binary latent ensemble, the framework constructs a hierarchy of tensors coupling measurement contexts to a latent Pascal organization through successive aggregation and folding operations. Each measurement context induces an observable partition of the same latent ensemble, and the resulting tensor formulation makes explicit the associated combinatorial multiplicities, shell organization, degeneracies, and induced reconstruction geometry. Rather than introducing a new reconstruction algorithm, this framework provides a mathematical representation of how latent configurations become organized under observation. The induced tensor kernel characterizes similarities between latent coordinates through their measurement-context responses, while regularized inversion provides a structured reconstruction of the latent representation from observable measurements. The binary construction further admits a natural multinomial extension to discrete simplex-supported latent representations. Connections to Pascal cellular automata and structured discrete color mappings illustrate respectively compressed and multinomial realizations of the framework. More generally, TENSKEL provides a combinatorial basis for reasoning about the organization induced by observation, with potential relevance to computational imaging, inverse problems, structured sensing, and quantum-inspired measurement formulations.

Related papers

• Connectome-to-Function: Conditional Generative Latent Representations for Reservoir Computing [3.376576692516218]
  We propose a conditional generative latent framework that encodes connectome graphs into a compact structural space.<n>From this space, the model can reconstruct observed connectivity with a mean edge-reconstruction AUC up to 0.910.<n>Using connectome-derived graphs as recurrent computational substrates, we found that the learned latent space captures functional variation across reservoir-computing experiments.
  arXiv  Detail & Related papers  (2026-09-05T13:39:08Z)
• Recurrent Neural Networks Beyond Time: Learning from Multiple Ordered Projections [13.839852906168787]
  We introduce the Ordered Structural Dependency Hypothesis (OSDH), which proposes that multiple admissible orderings of the same observations may reveal complementary structural dependencies inaccessible through a single sequential organization.<n>We present Structural Evolution RNNs (SE-RNNs), which employ conventional RNNs as projection-specific structural experts while preserving the underlying recurrent computation unchanged.
  arXiv  Detail & Related papers  (2026-08-10T14:56:20Z)
• CPrefix: A Combinatorial Tensor Framework for Structured Discrete Color Mappings [0.0]
  CPrefix is an observable representation for discrete mappings realized within a unified tensor framework.<n>The framework is validated on ICC display and printer profiles through latent reconstruction and perceptual transport.<n>It is applicable to structured multi-channel mappings arising from color imaging, spectral measurements and other discrete systems.
  arXiv  Detail & Related papers  (2026-08-04T16:04:27Z)
• PathAR: Structure-First Autoregressive Synthesis of Multimodal Pathology Images [51.428093790826814]
  We propose pathology Autorgressive modeling (PathAR) to factorize structure and appearance for modality-label-conditioned pathology generation.<n>PathAR stabilizes morphology under heterogeneous modality-specific appearances and enables spatially aligned image-mask pair generation.
  arXiv  Detail & Related papers  (2026-06-01T01:43:48Z)
• RiboSphere: Learning Unified and Efficient Representations of RNA Structures [57.40815107640066]
  RNA backbones are highly flexible, non-canonical interactions are prevalent, and experimentally determined 3D structures are comparatively scarce.<n>We introduce emphRiboSphere, a framework that learns emphdiscrete geometric representations of RNA by combining vector quantization with flow matching.
  arXiv  Detail & Related papers  (2026-03-20T04:36:35Z)
• The Representational Geometry of Number [1.5994376682356057]
  We show that number representations preserve a stable relational structure across tasks.<n>We find that task-specific representations are embedded in distinct subspaces, with low-level features like magnitude encoded along separable linear directions.<n>It suggests that understanding arises when task-specific transformations are applied to a shared underlying relational structure of conceptual representations.
  arXiv  Detail & Related papers  (2026-02-06T16:35:22Z)
• Geometric Decomposition of Statistical Inference through Gradient Flow and Co-Monotonicity Measures [0.0]
  We develop a geometric decomposition framework offering two strategies for partitioning inference problems into regional analyses.<n>We use path-monotonicity-validated discrete Morse theory to partition samples into gradient flow cells where outcomes exhibit monotonic behavior.<n>Co-monotonicity decomposition leverages association structure, defining embeddings of samples into association space.
  arXiv  Detail & Related papers  (2025-11-06T17:51:32Z)
• Compositional Structures in Neural Embedding and Interaction Decompositions [101.40245125955306]
  We describe a basic correspondence between linear algebraic structures within vector embeddings in artificial neural networks. We introduce a characterization of compositional structures in terms of "interaction decompositions" We establish necessary and sufficient conditions for the presence of such structures within the representations of a model.
  arXiv  Detail & Related papers  (2024-07-12T02:39:50Z)
• Bayesian Unsupervised Disentanglement of Anatomy and Geometry for Deep Groupwise Image Registration [59.062085785106234]
  This article presents a general Bayesian learning framework for multi-modal groupwise image registration.<n>We propose a novel hierarchical variational auto-encoding architecture to realise the inference procedure of the latent variables.<n>Experiments were conducted to validate the proposed framework, including four different datasets from cardiac, brain, and abdominal medical images.
  arXiv  Detail & Related papers  (2024-01-04T08:46:39Z)
• Reinforcement Learning from Partial Observation: Linear Function Approximation with Provable Sample Efficiency [111.83670279016599]
  We study reinforcement learning for partially observed decision processes (POMDPs) with infinite observation and state spaces. We make the first attempt at partial observability and function approximation for a class of POMDPs with a linear structure.
  arXiv  Detail & Related papers  (2022-04-20T21:15:38Z)
• Temporally-Coherent Surface Reconstruction via Metric-Consistent Atlases [131.50372468579067]
  We represent the reconstructed surface as an atlas, using a neural network. We empirically show that our method achieves results that exceed that state of the art in the accuracy of unsupervised correspondences and accuracy of surface reconstruction.
  arXiv  Detail & Related papers  (2021-04-14T16:21:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.