minor updates

Author: alegresor

index.md

@@ -7,8 +7,20 @@ layout: default
 
 # Softwares
 
-![image](./assets/points.svg)
 
+## FastGPs
+
+```
+pip install fastgps 
+```
+
+Gaussian process regression (GPR) models typically require $\mathcal{O}(n^2)$ storage and $\mathcal{O}(n^3)$ computations. [FastGPs](https://alegresor.github.io/fastgps)  implements GPR which requires only $\mathcal{O}(n)$ storage and $\mathcal{O}(n \log n)$ computations by pairing certain quasi-random sampling locations with matching kernels to yield structured Gram matrices. We support
+- GPU scaling,
+- batched inference,
+- robust hyperparameter optimization, and
+- multi-task GPR.
+
+![image](./assets/2d_gp.svg)
 
 ## QMCPy
 
@@ -18,10 +30,10 @@ pip install qmcpy
 
 [QMCPy](https://qmcsoftware.github.io/QMCSoftware/) is a Python package for Quasi-Monte Carlo (QMC) which contains
 - quasi-random (low discrepancy) sequence generators and randomization routines, including 
-    - **lattices** with
+    - *lattices* with
         - extensible constructions
         - random shifts
-    - **digital nets** (e.g. Sobol' points) with
+    - *digital nets* (e.g. Sobol' points) with
         - extensible constructions
         - random digital shifts,
         - linear matrix scrambling,
@@ -36,31 +48,17 @@ pip install qmcpy
 - a suite of diverse use cases, and
 - automatic variable transforms.
 
-
+![image](./assets/points.svg)
 
 ![image](./assets/mc_vs_qmc.svg)
 
-## FastGPs
-
-```
-pip install fastgps 
-```
-
-Gaussian process regression (GPR) models typically require $\mathcal{O}(n^2)$ storage and $\mathcal{O}(n^3)$ computations. [FastGPs](https://alegresor.github.io/fastgps)  implements GPR which requires only $\mathcal{O}(n)$ storage and $\mathcal{O}(n \log n)$ computations by pairing certain quasi-random sampling locations with matching kernels to yield structured Gram matrices. We support
-- GPU scaling,
-- batched inference,
-- robust hyperparameter optimization, and
-- multi-task GPR.
-
-![image](./assets/2d_gp.svg)
-
 ## QMCGenerators.jl
 
 ```
 ] add QMCGenerators
 ```
 
-[QMCGenerators.jl](https://alegresor.github.io/QMCGenerators.jl/) is a Julia package includes routines to generate and randomize quasi-random sequences used in Quasi-Monte Carlo. Supports the suite of low discrepancy sequence generators and randomization routines available in [QMCPy](https://qmcsoftware.github.io/QMCSoftware/). This package is a translation and enhancement of Dirk Nuyens' [Magic Point Shop](https://people.cs.kuleuven.be/~dirk.nuyens/qmc-generators/).
+[QMCGenerators.jl](https://alegresor.github.io/QMCGenerators.jl/) is a Julia package which includes routines to generate and randomize quasi-random sequences used in Quasi-Monte Carlo. This supports the suite of low discrepancy sequence generators and randomization routines available in [QMCPy](https://qmcsoftware.github.io/QMCSoftware/), see the description above. This package is a translation and enhancement of Dirk Nuyens' [Magic Point Shop](https://people.cs.kuleuven.be/~dirk.nuyens/qmc-generators/).
 
 
 

resume/sorokin_resume.pdf

@@ -26,21 +26,6 @@
 \social[github]{alegresor}
 \social[googlescholar]{akk3XSEAAAAJ}
 
-% \newcommand{\itlink}[2]{\textcolor{linkColor}{\textit{\link[#1]{#2}}}}
-% \newcommand{\bfitlink}[2]{\textcolor{linkColor}{\textbf{\textit{\link[#1]{#2}}}}}
-% \newcommand{\hpref}[2]{\hyperref[#1]{\textcolor{linkColor}{\textit{#2}}}}
-
-
-% \newcommand*{\newentry}[3][.25em]{%
-%   \begin{tabular}[t]{@{}p{0.125\textwidth}@{\hspace{\separatorcolumnwidth}}}%
-%     \raggedleft\hintstyle{#2}%
-%   \end{tabular}%
-%   \begin{tabular}[t]{@{}p{\dimexpr0.875\textwidth-\separatorcolumnwidth}@{}}
-%      #3
-%   \end{tabular}%
-%   \par\addvspace{#1}}
-
-
 \usepackage[
     backend=biber,
     %defernumbers=true,
@@ -74,8 +59,8 @@ \subsection{Education}
 \newentry{\normalfont{2017 - 2021}}{\textbf{B.S. in Applied Math, Minor in Computer Science.} IIT. Summa Cum Laude. GPA $3.94 / 4$.}
 
 \subsection{Experiences}
-\newentry{\normalfont{Jan - Dec 2025}}{\textbf{DOE SCGSR Fellow in Applied Mathematics} at \textbf{Sandia National Laboratory} in Livermore, CA. I am researching Gaussian process based scientific ML models for machine precision operator learning. I am also developing fast, scalable multi-task Gaussian processes for multi-fidelity modeling. We are preparing publications and open-source software with scalable GPU support e.g. see \texttt{FastGPs} below.}
-\newentry{\normalfont{Summer 2024}}{\textbf{Scientific Machine Learning Researcher} at \textbf{FM (Factory Mutual Insurance Company).} I built SciML models, including Physics Informed Neural Networks (PINNs) and Deep Operator Networks (DeepONets), for solving Radiative Transport Equations (RTEs) used to speed up CFD fire dynamics simulations. Resulted in publication of \citetitle{sorokin.RTE_DeepONet}.}
+\newentry{\normalfont{Jan - Dec 2025}}{\textbf{DOE SCGSR Fellow in Applied Mathematics} at \textbf{Sandia National Laboratory} in Livermore, CA. I am researching Gaussian process based scientific ML models for machine precision PDE solutions. I am also developing fast, scalable multi-task Gaussian processes for multi-fidelity modeling. We are preparing publications and open-source software with HPC support such as \texttt{FastGPs} below.}
+\newentry{\normalfont{Summer 2024}}{\textbf{Scientific Machine Learning Researcher} at \textbf{FM (Factory Mutual Insurance Company).} I built scientific ML models, including Physics Informed Neural Networks (PINNs) and Deep Operator Networks (DeepONets), for solving Radiative Transport Equations (RTEs) used to speed up CFD fire dynamics simulations. Resulted in publication of \citetitle{sorokin.RTE_DeepONet}.}
 \newentry{\normalfont{Summer 2023}}{\textbf{Graduate Intern} at \textbf{Los Alamos National Laboratory.} I modeled the solution processes of PDEs with random coefficients using efficient and error aware Gaussian processes. Resulted in publication of \citetitle{sorokin.gp4darcy}.}
 \newentry{\normalfont{Summer 2022}}{\textbf{Givens Associate Intern} at \textbf{Argonne National Laboratory}. I researched methods to efficiently estimate failure probability using Monte Carlo with non-parametric importance sampling. Resulted in publication of \citetitle{sorokin.adaptive_prob_failure_GP}.}
 \newentry{\normalfont{Summer 2021}}{\textbf{ML Engineer Intern} at \textbf{SigOpt, an Intel Company}. I developed novel meta-learning techniques for model-aware hyperparameter tuning via Bayesian optimization. In a six person ML engineering team, I contributed production code and learned key elements of the AWS stack. Resulted in publication of \citetitle{sorokin.sigopt_mulch}.}