About me
I'm an astrophysicist turned ML innovator. I design deep learning algorithms that efficiently extract meaning from complex, noisy data. I bring a scientist's rigor to every AI system I build.
I am also a passionate amateur astronomer and (astro-) photographer. I live in Munich, Germany.
Projects
-
LyαNNA: simulation-based inference with spectroscopic data in astrophysics
Improved posterior precision over conventional algorithms for intergalactic absorption spectra
-
SynTH: post processing hydrodynamic simulations
Modeling intergalactic absorption from massive cosmological hydrodynamic simulations and synthesizing mock spectra
-
GA-ILC: genetic algorithm for reconstructing the cosmic microwave background (CMB)
An evolutionary algorithm to extract faint cosmological signals from contaminated satellite observations (foreground separation)
Publications
-
Nayak, P.; M. Walther and D. Gruen (2025).
LyαNNA II: Field-level inference with noisy Lyα forest spectra.
arXiv e-prints, Oct 2025. https://arxiv.org/abs/2510.19899. -
Chang, S.; P. Nayak; M. Walther and D. Gruen (2025).
Human vs machine - 1:3. Joint analysis of classical and ML-based summary statistics of the Lyman-α Forest.
arXiv e-prints, Aug 2025. https://arxiv.org/abs/2508.03264. -
Nayak, P.; M. Walther; D. Gruen and S. Adiraju (2024).
LyαNNA: A Deep Learning Field-level Inference Machine for the Lyman-α Forest.
A&A, 689-A153, Sep 2024. https://doi.org/10.1051/0004-6361/202348485. -
Nayak, P. and R. Saha (2022).
Application of Genetic Algorithm to Estimate the Large Angular Scale Features of Cosmic Microwave Background.
MNRAS, 510-2, pp. 2173–2185, Feb 2022. https://doi.org/10.1093/mnras/stab3606.
Education
-
Ph.D. in Astrophysics, 2021–2025
LMU Munich, Germany
Thesis: Artificial intelligence for harvesting the untapped cosmological information in the Lyman-α forest. -
B.Sc + M.Sc. (Dual Degree) in Physics, 2015-2020
IISER Bhopal, India
Thesis: An application of genetic algorithm to estimate the large angular scale features of cosmic microwave background.
Get in touch
Feel free to drop me an email at parth3e8@gmail.com.
Connect with me on LinkedIn, Instagram or GitHub.
