PhD Defense by Fred Hohman

Event Details
  • Date/Time:
    • Monday October 5, 2020
      1:30 pm - 3:30 pm
  • Location: REMOTE: ZOOM
  • Phone:
  • URL:
  • Email:
  • Fee(s):
  • Extras:
No contact information submitted.

Summary Sentence: Interactive Scalable Interfaces for Machine Learning Interpretability

Full Summary: No summary paragraph submitted.

Title: Interactive Scalable Interfaces for Machine Learning Interpretability

Fred Hohman

School of Computational Science & Engineering
College of Computing
Georgia Institute of Technology

Date: Monday, October 5, 2020
Time: 1:30pm - 2:30pm EST
Location (remote via Zoom, password: fred):

Duen Horng (Polo) Chau - Advisor, Georgia Tech, Computational Science & Engineering
Alex Endert - Co-advisor, Georgia Tech, Interactive Computing

Chao Zhang - Georgia Tech, Computational Science & Engineering
Nathan Hodas - Pacific Northwest National Lab
Scott Davidoff - NASA Jet Propulsion Lab
Steven Drucker - Microsoft Research

Data-driven paradigms now solve the world's hardest problems by automatically learning from data. Unfortunately, what is learned is often unknown to both the people who train the models and the people they impact. This has led to a rallying cry for machine learning interpretability. But how do we enable interpretability? How do we scale up explanations for modern, complex models? And how can we best communicate them to people?

Since machine learning now impacts people's daily lives, we answer these questions taking a human-centered perspective by designing and developing interactive interfaces can enable interpretability at scale and for everyone. This thesis focuses on:

(1) Enabling machine learning interpretability: User research with practitioners guides the creation of our novel operationalization for interpretability, which helps tool builders design interactive systems for model and prediction explanations. We develop two such visualization systems, Gamut and TeleGam, which we deploy at Microsoft Research as a design probe to investigate the emerging practice of interpreting models. 

(2) Scaling deep learning interpretability: Our first-of-its-kind Interrogative Survey reveals critical yet understudied areas of deep learning interpretability research, such as the lack of higher-level explanations for neural networks. Through Summit, an interactive visualization system, we present the first scalable graph representation that summarizes and visualizes what features deep learning models learn and how those features interact to make predictions (e.g., InceptionNet trained on ImageNet with 1.2M+ images).

(3) Communicating interpretability with interactive articles: We use interactive articles, a new medium on the web, to teach people about machine learning's capabilities and limitations, while developing a new interactive publishing initiative called the Parametric Press. From our success publishing interactive content at scale, we generalize and detail the affordances of interactive articles by connecting techniques used in practice and the theories and empirical evaluations put forth by diverse disciplines of research.

This thesis contributes to information visualization, machine learning, and more importantly their intersection, including open-source interactive interfaces, scalable algorithms, and new, accessible communication paradigms. Our work is making significant impact in industry and society: our visualizations have been deployed and demoed at Microsoft and built into widely-used interpretability toolkits, our interactive articles have been read by 250,000+ people, and our interpretability research is supported by NASA.


Zoom Meeting Invitation

Topic: Fred Hohman's Dissertation Defense
Time: Oct 5, 2020 01:30 PM Eastern Time (US and Canada)

Join Zoom Meeting

Meeting ID: 899 6781 4013
Passcode: fred
One tap mobile
+13126266799,,89967814013#,,,,,,0#,,374057# US (Chicago)
+16465588656,,89967814013#,,,,,,0#,,374057# US (New York)

Dial by your location
        +1 312 626 6799 US (Chicago)
        +1 646 558 8656 US (New York)
        +1 301 715 8592 US (Germantown)
        +1 346 248 7799 US (Houston)
        +1 669 900 9128 US (San Jose)
        +1 253 215 8782 US (Tacoma)
Meeting ID: 899 6781 4013
Passcode: 374057
Find your local number:


Additional Information

In Campus Calendar

Graduate Studies

Invited Audience
Faculty/Staff, Public, Graduate students, Undergraduate students
Phd Defense
  • Created By: Tatianna Richardson
  • Workflow Status: Published
  • Created On: Sep 24, 2020 - 12:16pm
  • Last Updated: Sep 24, 2020 - 12:16pm