
Nathan Hogg
I am a Senior Honors student at Loyola University Chicago, majoring in Computer Science with a minor in Statistics. My research interests center on Computer Vision, 3D perception, and rendering, with a particular interest in how learning-based systems can reason about complex geometric and visual data. My current research focuses on robot perception, imitation learning, and 6D pose estimation. I am using modern vision foundation models including FoundationPose, SAM 3, and SAM 3D to track human and robotic motion, recover scene geometry, and explore motion retargeting from human demonstrations to robotic manipulators.Selected Work

An Adversarial Multi-Agent Pursuit Game for Locomotor Agility Pretraining
November 2026Humanoid robots learn to lunge, dodge, and dive by playing competitive tag against each other through league-based self-play.
MotionForge
August 2026 — PresentGPU-accelerated multi-agent adversarial pursuit games in simulation for pretraining agile legged locomotion policies.
KV Cache Eviction for Efficient LLM Inference
May 2026 — PresentDeveloping a vLLM KV-cache eviction policy to reduce memory pressure during large language model inference.
LeRobot SO-101 Arm
2026Built and programmed a LeRobot SO-101 robot arm from the open-source hardware design.

BELA: Learning Embodied Control Policies from Human Behavior
January 2026 — PresentLearning robotic control policies from human demonstrations using imitation learning and vision-language models.

Neural Scatter Removal for Image-Guided Radiation Therapy
May 2025 — August 2025A U-Net that removes megavoltage scatter from treatment images to support image-guided radiation therapy.
All projectsShow fewer
MoodMate
2025Full-stack mental wellness app offering AI-driven insights based on journal entries.
PlantPals
2025AI chatbot that teaches children about plant care as a Tamagotchi-like plant companion. Built at WildHacks.
Fair or Not? The Ethics of Unseen Data in AI Training
June 2024An analysis of consent, transparency, ownership, and fairness in large-language-model training data.

BitSource Crack Estimation
May 2024 — August 2024A hardware-software system that detects sidewalk cracks and estimates their severity with computer vision.

Determining Optimal Launch Conditions of Small-Scale Orbital Satellites
January 2022 — May 2022Mathematical modeling of the launch conditions needed to place small-scale satellites into orbit.
Experience
Research Assistant — LLM Inference Optimization
May — Aug 2026Argonne National Laboratory · Argonne Leadership Computing Facility
Lemont, IL
Developed a geometry-based KV-cache eviction method for vLLM to cut memory pressure in LLM inference. Modified vLLM internals and benchmarked latency and throughput at scale on Argonne systems, including the Aurora supercomputer. Owned the project end-to-end under Dr. Reet Barik. Funded by the DOE Research Assistant Program.
Research Assistant
2026 — PresentLoyola Chicago Software and Systems Laboratory
Chicago, IL
Robot learning research with Matt Hyatt, Dr. Neil Klingensmith, and Dr. George Thiruvathukal. On the BELA project I program and run manipulation experiments on a UFACTORY xArm, teleoperate it with a GELLO leader arm, and built a SAM 3 / SAM 3D Objects perception pipeline for 3D meshes and 6-DoF object state. Also lead MJTag / MotionForge, multi-agent adversarial games for agile humanoid locomotion.
Computer Science Tutor
2026 — PresentLoyola University Chicago
Chicago, IL
Tutor students in Intro to Computing, OOP, Discrete Structures, Data Structures, and Visual Information Processing.
Webmaster
2025 — PresentLoyola University Chicago — Dept. of Computer Science
Chicago, IL
Maintain the department website with 99.9% uptime, collaborate with faculty on new features, and improve performance and accessibility.
Research Intern — Radiation Oncology
May — Jul 2025Loyola Stritch School of Medicine, Cardinal Bernardin Cancer Center
Maywood, IL
Trained a U-Net to remove MV scatter from kV images acquired during lung-cancer radiation therapy, improving PSNR 46%, SSIM 27%, and MAE 73% and generalizing to unseen rotational scans — a proof of concept for AI-assisted tumor tracking.
Software Development Intern
May — Sep 2024Bit Source
Pikeville, KY
Built an automated sidewalk-scanning system fusing an Intel RealSense camera and GNSS on a Raspberry Pi, with a scooter-mounted rig and touchscreen UI. Also built an OCR pipeline that digitizes physical newspapers into a Pinecone vector database.
Research Assistant — Moseley Bioinformatics Lab
2022 — 2024University of Kentucky Markey Cancer Center
Lexington, KY
Processed and analyzed Fourier transform mass spectrometry data and investigated methods to minimize sinc-wiggle distortions.
Show earlier experienceHide earlier experience
Lab Member
2023Abegaz Research Group
Chicago, IL
Built a custom robotic platform and developed MATLAB machine-learning algorithms for real-time object detection.
Software Engineering Intern
2022 — 2024Bit Source
Remote / Pikeville, KY
Built a full-stack sports-league platform for Appalachian coal-camp communities (2023–24) and a stablecoin price and trend dashboard, taking it from customer requirements to delivery (2022).
Researcher — ACTION Program
2020 — 2022University of Kentucky Markey Cancer Center
Lexington, KY
Cancer research, clinical, and outreach training program for Appalachian Kentucky students; contributed to three published student collections.
Publications
- An Adversarial Multi-Agent Pursuit Game for Locomotor Agility Pretraining
N. A. C. Hogg, M. Hyatt, N. Klingensmith, G. K. Thiruvathukal
CoRL 2026 · RoboLetics 3.0 WorkshopRunner-up, Best Paper Behavior is a Language for Cross-Embodiment Control
Third author
IROS 2026 · Poster
How the Coronavirus Is Widening the Digital Divide: A Student Perspective
Kentucky Student Voice Forumn
Cancer in Appalachia: A Collection of Youth-Told Stories, Volume One
ACTION Program, UK Markey Cancer Center
The Cancer Crisis in Appalachia: Kentucky Students Take ACTION, 2nd Edition
ACTION Program, UK Markey Cancer Center
Funding
- Emergent Locomotor Agility Through Multi-Agent Adversarial Games
Department Research Funding $5,000
2026 - Learning Embodied Control Policies from Human Behavior
Provost Fellowship $2,500
2026 - Neural Network to Reduce MV Scatter in kV Images
Summer Research Fellowship $6,400
2025
Education
Loyola University Chicago
B.S. Computer Science, Minor in Statistics · Interdisciplinary Honors
Heriot-Watt University
Visiting Student, Computer Science
The Gatton Academy at Western Kentucky University
Mathematics & Computer Science coursework; dual enrollment
Morehead State University
Mathematics & Computer Science coursework; dual enrollment
Rowan County Senior High School
High School Diploma
Honors & Awards
- Provost FellowshipLoyola University Chicago
- Dean's ListLoyola University Chicago
- Dean's ListWestern Kentucky University
- Presenter, Data Science Half-DayWestern Kentucky University, 2022
- Presenter, Mathematics SymposiumWestern Kentucky University, 2023
Skills
Contact
I'm always happy to talk about research collaborations, internships, or interesting problems. The best way to reach me is by email.
nathanhogg1223 [at] gmail [dot] com