# Mathscapes Applied machine learning, research tooling, and technical figures for research and engineering teams. Website: https://mathscapes.xyz Contact: hi@gvsh.cc ## Services ### Applied ML & research tooling Models and analysis pipelines for engineering data, delivered as documented, tested software your team can keep. ### Technical figures & communication Diagrams, mathematical explanations and data visualizations for papers and reports, built with editable source. ## Selected work ### Classifying auxetic deformation Follow a re-entrant honeycomb from simulation and plastic-hinge images to cluster labels, then examine what the reported classifier agreement can tell us. PDF: https://mathscapes.xyz/assets/portfolio/from-simulation-to-classification.pdf Details: https://mathscapes.xyz/portfolio/from-simulation-to-classification/ Source: https://doi.org/10.1177/20414196241281069 Scope: An original research interpretation of the accepted manuscript. The specimen and image-processing strip are explanatory illustrations, not original simulation output. Reported results are distinguished from a proposed evaluation. This is one of two portfolio perspectives on the same paper. ### Lazy evaluation for streaming statistics Trace lazy filtering and a running mean through six source values, then compare reported throughput when a consumer stops early. PDF: https://mathscapes.xyz/assets/portfolio/statistics-without-the-history.pdf Details: https://mathscapes.xyz/portfolio/statistics-without-the-history/ Source: https://doi.org/10.5281/zenodo.18610143 Scope: The source trace is a computed six-value example. The throughput chart reproduces reported manuscript benchmarks; these were not rerun. This is one of four portfolio perspectives on the same paper. ### Anomaly scoring before the baseline update A single-page research infographic with the study context, methods, evidence and limitations. Why the ordering of an online anomaly detector changes its answer. PDF: https://mathscapes.xyz/assets/portfolio/score-before-you-update.pdf Details: https://mathscapes.xyz/portfolio/score-before-you-update/ Source: https://doi.org/10.5281/zenodo.18610143 Scope: This is a worked arithmetic example, not an anomaly benchmark or a fitted detector. This is one of four portfolio perspectives on the same paper. ### Deformation modes in eight image clusters Map eight image clusters to four named deformation mechanisms and three developing states, with unrecovered reference modes identified. PDF: https://mathscapes.xyz/assets/portfolio/reading-deformation-modes.pdf Details: https://mathscapes.xyz/portfolio/reading-deformation-modes/ Source: https://doi.org/10.1177/20414196241281069 Scope: An original crosswalk of reported category relationships. Small motifs are mnemonic labels, not recovered meshes or stress and strain fields. This is one of two portfolio perspectives on the same paper. ### Updating variance without storing the data A geometric derivation of the online variance update, using [4, 7, 5] and exact squared-deviation areas. PDF: https://mathscapes.xyz/assets/portfolio/anatomy-of-online-variance.pdf Details: https://mathscapes.xyz/portfolio/anatomy-of-online-variance/ Source: https://doi.org/10.5281/zenodo.18610143 Scope: An original mathematical explanation using [4, 7, 5]. The underlying recurrence is documented in the Iterflow manuscript. This is one of four portfolio perspectives on the same paper. ### EWMA response, memory and noise Compare EWMA step response, past-input weights, settling time and theoretical noise attenuation. PDF: https://mathscapes.xyz/assets/portfolio/the-memory-of-a-smoother.pdf Details: https://mathscapes.xyz/portfolio/the-memory-of-a-smoother/ Source: https://doi.org/10.5281/zenodo.18610143 Scope: Computed response to a step from 2 to 8, shown over 16 updates. Weight strips describe infinite steady-state history; noise ratios assume independent input noise. This is one of four portfolio perspectives on the same paper. ## Publications ### 2026 — A framework for visual-mathematical literacy in applied machine learning: why representation choices shape what models can learn Gaurav Singh, R. S. Dhari. Machine Learning: Engineering (IOP Publishing). https://doi.org/10.1088/3049-4761/ae7df3 ### 2026 — Iterflow: Composable Streaming Statistics for JavaScript Gaurav Singh. Mathscapes (Zenodo). https://doi.org/10.5281/zenodo.18610143 ### 2024 — Automated detection of deformation mechanisms in re-entrant honeycomb auxetics using machine learning Gaurav Singh, R. S. Dhari, Zia Javanbakht. International Journal of Protective Structures. https://doi.org/10.1177/20414196241281069 ## People Gaurav Singh Mathscapes https://mathscapes.xyz/people/gaurav-singh/ Rahul Singh Dhari Curtin University, Australia https://mathscapes.xyz/people/rahul-singh-dhari/