Nick Sohre

PhD Candidate, University of Minnesota CSE

My goal is to help empower humanity by creating more social forms of artificial intelligence that incorporate an understanding of human activity into their behavior. My research interests include Machine Learning and Data-Driven AI, Graphics, and VR.

CV • LinkedIn

Recent Publications

Dynamic Properties of Successful Smiles (image missing)
PLOS ONE 2017 Research Article

Dynamic Properties of Successful Smiles

Facial expression of emotion is a foundational aspect of social interaction. We use a computer animated 3D facial tool to systematically produce a range of different smiles, and ask 802 participants to rate...

Full Report (PDF)
Implicit Crowds: Optimization Integrator for Robust Crowd Simulation (image missing)
SIGGRAPH 2017 Full Paper

Implicit Crowds: Optimization Integrator for Robust Crowd Simulation

Large multi-agent systems involve interactions that are anticipatory in nature. We propose a simple and effective optimization-based integration scheme for the implicit integration of such systems...

Full Report (PDF)Watch Video
Evaluating Collision Avoidance Effects on Discomfort in Virtual Environments (image missing)
IEEE VR 2017 VHCIE Workshop Paper

Evaluating Collision Avoidance Effects on Discomfort in Virtual Environments

Here, we explore the role collision avoidance between virtual agents and the VR user plays on experiences in an immersive virtual environment. When Collision avoidance was used, we found...

Full Report (PDF)
Data Driven Sokoban Puzzle Generation with MCTS (image missing)
AIIDE 2016 Full Paper (Best Student Paper Award)

Data Driven Sokoban Puzzle Generation with MCTS

In this work, we propose a Monte Carlo Tree Search (MCTS) based approach to procedurally generate Sokoban puzzles. Our method generates puzzles through simulated game play, guaranteeing solvability...

Full Report (PDF)

Research

My work seeks to combine state of the art AI with real world data to understand and solve complex problems

  • Using Data to Understand and Utilize Human Motion in Medicine
    We work with collaborators in Quantitative Psychology and Facial Surgery to discover how human motion can inform and improve medical procedures, training, and therapy.
  • Adaptive PCG for Games
    We combine artificial intelligence algorithms with large scale datasets to adaptively tune gameplay, allowing for user-oriented experiences that transcend user skill level or game familiarity.
  • Exploring Interactions in Virtual Reality
    Using computer vision techniques, we work to capture, recognize, and utilize human activity to provide more comfortable, immersive, and engaging experiences in virtual environments.
  • Stable, Robust, and High Quality Multi-Agent Simulation
    We work to improve our ability to model how humans move through spaces. The ability to accurately simulate human navigation is key for a myriad of applications, including robotics, architecture, and animation.

Education

University of Minnesota

Dordt College

Experience

My professional experience includes teaching, software engineering, and development on both large and small scale projects.

University of Minnesota

IBusiness Solutions, Inc