MILIND
SAHU_
Mathematics · Computing · Energy Systems
01 About
I'm a BS-MS Mathematics student at IISER Tirupati (Indian Institute of Science Education and Research, an autonomous research institute of the Govt. of India), working at the intersection of mathematical modeling, optimization theory, and computational methods applied to real-world systems.
I hold research positions across three institutes — a Gaussian process regression fellowship at IIT Palakkad, a research collaboration with Dr. Sven Karbach at the University of Amsterdam on Wasserstein-space forecasting, and a diffusion-geometry proposal at IIIT Kottayam — and I have two published papers: arbitrage-free GP interpolation of implied volatility surfaces, and a policy and ML framework for civic energy disaggregation.
Through my research identity AXL — Axiom Labs, I also build applied systems — from NILM-based energy disaggregation to proof-theoretic waste traceability — and compete in national hackathons at IITs and IISERs, placing in the top ranks against thousands of participants.
IISER Tirupati — BS–MS Dual Degree, Major: Mathematics · 2025–Present (Expected 2030)
Coursework: Real Analysis, Linear Algebra, Probability Theory, Convex Optimization, Stochastic Processes, Time-Series Analysis
Competitive entry: AIR 780 — NEST (National Entrance Screening Test) · AIR 3714 — IAT (IISER Aptitude Test)
Outside research — I shoot and edit. Mostly street, campus life, and the quiet moments between things. @its_milindsahu →
02 Research & Publications
Gaussian Process Regression for Distribution-Valued Inputs
Replicates Bachoc et al. (2018) on GP regression with probability distributions as inputs, then extends it to Wasserstein Tangent Gaussian Processes.
Open reader →Renewable Capture-Price Forecasting with Wasserstein GPs
Treats the joint price–production forecast as a probability measure so that the cannibalisation effect in renewable markets enters the GP.
Open reader →Wasserstein Diffusion on the Annotation Simplex
Corrects the geometry of diffusion-based video summarisation: barycenter consensus, Dirichlet noise and a Wasserstein loss.
Open reader →Arbitrage-Free Gaussian Process Interpolation for Implied Volatility Surfaces
Milind Sahu, Darun, Latheesh Roy
A GP-based interpolation scheme for implied volatility surfaces that embeds no-arbitrage constraints directly in the covariance structure, avoiding post-hoc smoothing or projection.
DOI 10.13140/RG.2.2.33170.62400 ↗From Opt-In to Infrastructure: Non-Intrusive Load Monitoring, Sustained Household Energy Conservation, and the Civic Energy Disaggregation Mandate
Milind Sahu, Darun A. · Axiom Lab · CC BY-NC 4.0
Introduces the Civic Energy Disaggregation Mandate (CEDM), arguing for NILM as public utility infrastructure rather than opt-in consumer technology, with a five-model ML pipeline benchmarking disaggregation accuracy and deployment feasibility.
DOI 10.13140/RG.2.2.20940.17283 ↗03 Research Galaxy
— Certificates & Awards
04 Research Identity
Axiom Labs is my independent computational research identity — building at the intersection of rigorous mathematics and deployable systems. Every project starts from axioms, not assumptions.
Explore AXL →05 Projects
11 Awards · IITs & IISERs · 5000+ Competitors
06 Gallery
