CV

Machine Learning Engineer

Ultra Electronics

Developed models to reduce operator cognitive load, and increase battlespace understanding.

Student Researcher

University of New Hampshire

Investigated connections between inverse reinforcement learning, behavioral cloning and maximum entropy methods.

Database and API Developer

Brands Express LLC

Developed and maintained database and API integrations with ecommerce platforms.

PhD in Computer Science

University of Colorado Boulder

Department of Computer Science

Bachelor of Science

University of New Hampshire

Department of Mathematics and Statistics

Reinforcement learning for Partially Observed Markov Decision Processes

Traditional POMDP solves do not scale to realistic problems. POMDPs also violate the assumption of existing RL framework, therefore we need a representation that allows existing RL to bne used for POMDPs.

• Reinforcement Learning
• Partial Observed Markov Decision Processes

Semantic Octree Latent Diffusion

Latent diffusion models are a powerful tool for generating high-quality images. Use representaitons like octrees we can generate semantically rich 3d scenes for robotapplications.

• Latent Diffusion
• Semantic Scene Generation
• 3D Scene Generation

Robot Centric Foundation Models

Use rich sensor data to train foundation models that can be used for a variety of robot tasks.

• Robot Learning
• Foundation Models
• Robot Navigation