Update index.html

Author: Hariprashad-Ravikumar

index.html

@@ -206,7 +206,7 @@ <h1>Hariprashad Ravikumar</h1>
 </p>
 
 <p class = "a">
-My PhD research under <a href="https://phys.nmsu.edu/facultydirectory/engelhardt_michael.html" target="_blank">Dr. Michael Engelhardt</a> (NMSU) focuses on lattice quantum chromodynamics (QCD) calculations of Transverse Momentum Dependent Parton Distribution Functions (TMDs). To achieve this, I built an end-to-end machine learning pipeline to process over 30,000 multidimensional observables from Monte Carlo simulations, achieving <strong>93%+ model fit accuracy</strong> using symbolic regression (PySR) with physics-constrained loss functions. To handle the multi-terabyte datasets generated, I developed GPU-accelerated CUDA C++ (cuFFT) pipelines, which <strong>reduced data processing time by 10x</strong> on HPC clusters. For robust analysis, I also created production-grade Python and Mathematica packages to manage jackknife resampling and ensure numerical stability.
+My PhD research under <a href="https://phys.nmsu.edu/facultydirectory/engelhardt_michael.html" target="_blank">Dr. Michael Engelhardt</a> (NMSU) focuses on lattice quantum chromodynamics (QCD) calculations of Transverse Momentum Dependent Parton Distribution Functions (TMDs). To achieve this, I built an end-to-end machine learning pipeline to process over 30,000 multidimensional observables from Monte Carlo simulations, achieving <strong>98%+ model fit accuracy</strong> using symbolic regression (PySR) with physics-constrained loss functions. To handle the multi-terabyte datasets generated, I developed GPU-accelerated CUDA C++ (cuFFT) pipelines, which <strong>reduced data processing time by 10x</strong> on HPC clusters. For robust analysis, I also created production-grade Python and Mathematica packages to manage jackknife resampling and ensure numerical stability.
 </p>