Siddharth Mishra-Sharma's Projects
AI-powered arXiv research assistant prototype.
Angular power spectra for astrometric weak lensing. Code repository associated with the paper https://arxiv.org/abs/2003.02264.
Community-sourced list of papers and resources on neural simulation-based inference.
A cybernetic physicist sent back in time to fuse techniques from quantum field theory and machine learning. April Fools' 2023.
Public repository of the Cosmic Linear Anisotropy Solving System (master or 2.7 branches for standard code; ExoCLASS branch for exotic energy injection; class_master branch for FFTlog)
Self-contained example of differentiable CMB parameter inference with power spectrum emulators.
Implementation of Consistency Models (Song et al 2023) for few-step image generation in Jax.
LaTeX CV.
Dark photon conversions in our inhomogeneous Universe. Code repository associated with the papers https://arxiv.org/abs/2002.05165 and https://arxiv.org/abs/2004.06733.
Code implementing event generation and search strategies for semi-visible jets. Associated to 1707.05326.
Tutorials for the 2022 IAIFI Summer School, covering (deep) probabilistic programming with Jax and NumPyro.
Code for dark matter likelihood scans on gamma-ray data. Repository associated with 1708.09385 and 1709.00416.
Studying the effects of resonant photon injection in 21-cm observations. Code repository associated with the paper https://arxiv.org/abs/2009.03899.
Simulation-based priors connecting EFT of large-scale structure and halo occupation distribution (HOD) models.
Annotated implementations of equivariant neural networks in Jax.
Simulation-based inference for the Galactic Center gamma-ray excess. Code repository associated with https://arxiv.org/abs/2110.06931.
Proof-of-principle application of Gaussian process modeling to gamma-ray analyses. Code repository associated with the paper https://arxiv.org/abs/2010.10450.
p(spectra|light curves)
Multi-probe Fisher code
Basic Gaussian processes on a sphere. Written at the Gaia Sprint 2019 (KITP).
Hierarchical neural implicit inference over event ensembles. Code repository associated with https://arxiv.org/abs/2306.12584.
Starter notebooks for self-supervised/multimodal machine learning and neural simulation-based inference, for the 2024 IAIFI Astro Hackathon.
Normalizing flow models allowing for a conditioning context, implemented using Jax, Flax, and Distrax.
A Jax wrapper for cudaKDTree
Neural Simulation-based Inference with GNN for Jeans Modeling
Set-based generative modeling of particle jets using continuous normalizing flows.
Implicit neural representations for strong lensing source reconstruction. Code repository associated with https://arxiv.org/abs/2206.14820.
Bare-bones implementations of some generative models in Jax: diffusion, normalizing flows, consistency models, flow matching, (beta)-VAEs, etc