Hello, my name is Sonny! 🐣

About Me 📰

As a passionate data-driven researcher, I thrive on collaboration and innovation, particularly in leveraging machine learning to revolutionize healthcare. I contribute to cutting-edge projects at the NERVES Lab and the Utah NeuroRobotics Lab at the University of Utah. My work included designing wireless sEMG chips, programs to process body-worn sensor signals automatically, and developing cognitive workload measurements for complex machine learning systems for neuroprosthesis.

I am working to create advanced learning systems that use state-of-the-art reinforcement learning algorithms to control exoskeletons for post-stroke assistance while walking on and between different terrains. I seek opportunities to develop my data science, artificial intelligence, and machine learning expertise and drive impactful research through collaborative endeavors.

Let’s innovate together! #ML #AI # RL#DataScience

I'm currently pursuing my Ph.D in Biomedical Engineering with an emphasis in Data Science and Computation 💻
I'm currently a Graduate Research Assistant for the NERVES Lab, University of Utah 🧠
Appropriate coursework: Data Science for Biomedical Engineers, Reinforcement Learning Specialization, Machine Learning, Artificial Intelligence, Deep Learning, NeuroRobotics
Interested in Machine Learning, Reinforcement Learning, Deep Learning, and AI 🦾

Programming Skills

Tools and Software

Education 📖

Biomedical Engineering, PhD @ University of Utah (In-Progress)

Biomedical Engineering, MS @ University of Utah (In-Progress)

Biomedical Engineering, BS @ University of Utah

Certification 📃

Graduate Certificate in Deep Learning (In-Progress) (University of Utah)

[Certification Homepage]

Reinforcement Learning Specialization (University of Alberta & Alberta Machine Intelligence Institute)

[Specialization Homepage][View Certificate]

Data Science Professional (DataCamp)

[Course Homepage][View Certificate]

Machine Learning Scientist (DataCamp)

[Course Homepage][View Certificate]

Research Experience 🥼

Graduate Research Assistant @ NERVES Lab

Research Assistant @ NeuroRobotics Lab

Research Analyst/Assistant @ TORCH

Undergraduate Research Assistant @ Center for Neural Interfaces

Projects 💻

Fine-Tuning Faster R-CNN for Guardrail Damage Detection [View Project]

Researched and implemented advanced fine‑tuning strategies on a Faster R‑CNN to detect guardrail damage, improving detection accuracy by 5% and decreasing false positives by 92% for Blyncsy, Inc.
Designed a saliency‑scoring pipeline leveraging pre‑trained networks to automatically verify ground truth labels in the training and testing sets, automatically finding errors in roughly 10% of images.
Engineered end‑to‑end training pipelines for a custom Faster R‑CNN, packaged as the BlyncsySFT pip package for easy deployment and reproducibility.
Keywords: Python, PyTorch, Object Detection, pip

Machine Learning Library From Scratch [View Project]

Built a custom Python ML library with cross‑validation to support ID3, perceptron, logistic‑regression, and SVM classifiers.
Leveraged library to secure a top‑10 finish out of 150+ teams in a course Kaggle NLP classification challenge.
Performed feature transformation, dimensionality reduction, and hyperparameter tuning to increase model accuracy.
Algorithms: ID3 decision tree, mistake‑bound perceptron, MLE/MAP logistic regression, objective‑minimization SVM, ensemble methods (bagging, AdaBoost, SVM‑over‑trees)
Keywords: Python, Decision Tree, Perceptron, Logistic Regression, SVM, Ensembles

Helping Pac-Man Using AI Techniques [View Project]

Implemented various search algorithms (A*, Alpha-Beta, Expectimax) to solve navigation problems efficiently.
Utilized RL methods to teach Pac-Man how to navigate the grid world environment and chase ghosts.
Created probabilistic models using Bayes Nets/Particle Filtering to help Pac-Man hunt invisible ghosts from observations.
Breakdown of Projects:
- Project 1: Search. BFS, DFS, A*
- Project 2: Minimax, Alpha-Beta, Expectimax
- Project 3: RL, Value and Policy Iteration, Q-Learning
- Project 4: Bayes Nets, Particle Filtering
Keywords: Python, Artificial Intelligence (AI), Reinforcement Learning (RL), Search, Bayes Nets

Lunar Lander Reinforcement Learning Agent [View Project]

Implemented reinforcement learning algorithms to control agents to land on a simulated moon in OpenAI Gym.
Trained deep Q-Learning network with RMSProp to control action selection.
Explored the effects of different learning parameters on agent success and behavior.
Software: Python, Jupyter, OpenAI Gym
Keywords: Reinforcement Learning (RL), Function Approximation, SARSA, Q-Learning

Singular Value Decomposition (SVD) for Trend Identification in Valorant Ranked Data [View Project]

Utilized SVD to identify trends in player statistics for Valorant.
Identified multiple significant statistics that might help differentiate players by rank.
Developed an NN (Multilayer Perceptron) classifier to predict player rank from player statistics.
Software: Python, Jupyter, Sklearn, SciPy
Keywords: SVD, MLP

Predicting Property Rental Prices [View Project]

Developed machine learning models to predict property rental prices from housing data.
Utilized exploratory methods like correlation matrices and scaling for feature engineering.
Implemented stacking KNN and regression techniques to improve model performance.
Software: Python, Jupyter, Sklearn
Keywords: Modeling, Regression, Decision Trees, KNN

Predicting Credit Card Approvals [View Project]

Developed machine learning models based on previous approval data to predict credit card approval.
Software: Python, Jupyter, Sklearn
Keywords: Modeling, Regression