Jace Robinson
CV (Last updated Dec 2017): [pdf]
LinkedIn: [Link]
Google Scholar: [Link]
Github: [Link]

I am currently a graduate research assistant in the Web and Complex Systems Lab and Kno.e.sis Research Center in the computer science department at Wright State University, working towards a Master's of Science in Computer Science under the advisement of Dr. Derek Doran. I am fascinated with the notation of mapping mathematical and statistical ideas to real data with impact. Currently I am interested in combining ideas from machine learning, network science, and time series to tackle problems in complex systems. I will be pursuing a PhD with a focus in network science and machine learning starting Fall 2018.

I am actively looking for summer internship in data science or data analyst related positions for 2018! Feel encouraged to contact me at robinson.329@wright.edu with opportunities.

Dynamic Network Modeling
Dynamic Network Modeling
Sociotechnological and geospatial processes exhibit time varying structure that make insight discovery challenging. This project proposes a new statistical model for such systems, modeled as dynamic networks, to address this challenge. It assumes that vertices fall into one of k types and that the probability of edge formation at a particular time depends on the types of the incident nodes and the current time. The time dependencies are driven by unique seasonal processes, which many systems exhibit (e.g., predictable spikes in geospatial or web traffic each day). The project defines the model as a generative process and an inference procedure to recover the seasonal processes from data when they are unknown.

So far we have one workshop publication [1] with more to come!
SDSBM Overview Figure

Probabilistic Risk Assessment for Astronauts
In the past summer of 2017, I worked as an intern at the National Aeronautics and Space Administration (NASA) Glenn Research Center with Debra Goodenow and Jerry Myers. I worked in the human research program to develop a dynamic probabilistic risk assessment tool to quantify the level of risk astronauts face from medical conditions in various mission scenarios.

We expect to publish at the NASA Human Research Program Investigators' Workshop January 2018!

Parallel Iterative Closest Point
Iterative Closest Point for Automated Aerial Refueling
The Iterative Closest Point algorithm is a widely used approach to aligning the geometry between two 3 dimensional objects. The capability of aligning two geometries in real time on low-cost hardware will enable the creation of new applications in Computer Vision and Graphics such as automated aerial refueling between a tanker and jet. This work presents an accelerated alignment variant which utilizes parallelization on a Graphics Processing Unit (GPU) of multiple kNN approaches augmented with a novel Delaunay Traversal to achieve real time estimates.

This work resulted in two conference publications [2] [3].

Nonlinear Optimization
Nonlinear Optimization with Confidence Visualization for Tracking Passive Objects
Angle of Arrival (AOA) measurements can be created from any sensor platform with any sort of optical sensor, location and attitude knowledge to track passive objects. Previous work has created a non-linear optimization (NLO) method for calculating the most likely estimate from AOA measurements. Two new modifications to the NLO algorithm are created and shown to correct AOA measurement errors by estimating the inherent bias and time-drift in the Inertial Measurement Unit (IMU) of the AOA sensing platform.

This work resulted in a conference publication [4].

[1] Robinson J., Doran D., “Seasonality in Dynamic Stochastic Block Models", Proceedings of the International Conference on Web Intelligence, ACM, 2017. [Link] [arXiv] [slides]

[2] Robinson J., Piekenbrock M., Burchett L., Nykl S., Woolley B., Terzuoli A., “Parallelized Iterative Closest Point for Autonomous Aerial Refueling”, Proceedings of Advances in Visual Computing (Lecture Notes in Computer Science 10072), 2016. [Link]

[3] Burchett L., Robinson J., Piekenbrock M., Nykl S., Woolley B., and Terzuoli T., “Automated aerial refueling: Parallelized 3d iterative closest point”, Proceedings in IEEE (International). Conference in Aerospace & Electronics, 2016. [Link]

[4] Levy D., Roos J., Robinson J., Carpenter W., Martin R., Taylor, C., Sugrue J., Terzuoli A., “Non Linear Optimization Applied to Angle-Of-Arrival Satellite Based Geo-Localization for Biased and Time-Drifting Sensors”, In International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences 41, 2016. [Link]

[5] Robinson J., “Investigation of Algebraic Combinatorics through Difference Sets”, Undergraduate Thesis, Wright State University, 2016.

[6] Phillips B., Robinson J., “Some New Almost Difference Sets Via Finite Fields”, ACM Communications in Computer Algebra 49, 2015. [Link]