Marko Sterbentz

Marko Sterbentz

Artificial Intelligence PhD Candidate

Northwestern University

C3 Lab

I am a computer science PhD candidate at Northwestern University in the C3 Lab advised by Kristian Hammond. My research is centered around automating data science processes in order to enable people to directly ask questions of their data and receive meaningful insights and information in return. In general, I am interested in developing and applying methods which combine neural and symbolic methods in order to improve access to data and information. In particular, I’ve been developing a neurosymbolic method which identifies user information needs, formulates and executes analytic plans to derive this information from data, and then communicates this information with highly fluent language via a large language model. The goal of this work is to produce factual and fluent documents which communicate meaningful information from a person’s data.

I am also affiliated with Northwestern’s Center for Advancing Safety in Machine Intelligence (CASMI) and am interested in developing methods for ensuring machine learning systems are safe and effective in the real-world.

Interests

  • Artificial Intelligence
  • Question Answering
  • Neurosymbolic AI
  • Language Generation

Education

  • PhD in Computer Science, 2019 - present

    Northwestern University

  • MS in Computer Science, 2019

    University of Southern California

  • BS in Computer Science, 2017

    Idaho State University

News

Publications

Lightweight Knowledge Representations for Automating Data Analysis

The principal goal of data science is to derive meaningful information from data. To do this, data scientists develop a space of …

Summarization from Leaderboards to Practice: Choosing A Representation Backbone and Ensuring Robustness

Academic literature does not give much guidance on how to build the best possible customer-facing summarization system from existing …

Requirements for Open Political Information: Transparency Beyond Open Data

A politically informed citizenry is imperative for a well-developed democracy. While the US government has pursued policies for open …

From Data to Information: Automating Data Science to Explore the US Court System

The U.S. court system is the nation’s arbiter of justice, tasked with the responsibility of ensuring equal protection under the …

GPGPU Enabled Ray Directed Adaptive Volume Visualization for High Density Scans

This paper presents an open source implementation of a volume visualizer capable of rendering large scale simulated and tomographic …

Experience

 
 
 
 
 

PhD Candidate

Northwestern University

Sep 2019 – Present Illinois
 
 
 
 
 

Research Intern

Lawrence Livermore National Laboratory

May 2019 – Aug 2019 California
 
 
 
 
 

Research Intern

Idaho National Laboratory

May 2018 – Aug 2018 Idaho

Teaching

Teaching Assistant

  • NU - CS 338: Practicum in Intelligent Information Systems
    • Fall 2022
    • Spring 2022
    • Winter 2022
    • Fall 2020
  • USC - CSCI 576: Multimedia Systems Design
    • Spring 2019
    • Fall 2018