event

PhD Defense by Subhajit Das

Primary tabs

Title

EMPOWERING USERS COMMUNICATE THEIR PREFERENCES TO MACHINE LEARNING MODELS IN VISUAL ANALYTICS

 

Subhajit Das

PhD Candidate in Computer Science

School of Interactive Computing

Georgia Institute of Technology

 

Date: Friday, November 20, 2020

Time: 4:00pm - 6:00pm EST (virtual)

Location: (remote via MS Teams, see below) 

 

Committee

Dr. Alex Endert - Advisor, Georgia Institute of Technology, School of Interactive Computing

Dr. John Stasko -  Georgia Institute of Technology, School of Interactive Computing

Dr. Polo Chau - Georgia Institute of Technology, School of Interactive Computing

Dr. Thomas Ploetz - Georgia Institute of Technology, School of Interactive Computing

Dr. Remco Chang - Tufts University, School of Computer Science

 

Abstract

Recent visual analytic (VA) systems rely on machine learning (ML) to allow users perform a variety of data analytic tasks, e.g., biologists clustering genome samples, medical practitioners predicting the diagnosis for a new patient, ML practitioners tuning models' hyperparameter settings etc. These VA systems support interactive construction of models to people (I call them power users) with a diverse set of expertise in ML; from non-experts, to intermediates, to expert ML users. However, designing VA systems for power users poses various challenges, such as addressing diversity in user expertise, metric selection, user modeling to automatically infer preferences, evaluating the success of these systems, etc. Through my research, I investigate how power users can communicate their preferences to interactively construct machine learning models in support of various data analytic goals.

Specifically, I explore and analyze several VA techniques such as multi-model steering, model selection, interactive objective functions, conflict resolution in objective functions etc. to facilitate specification of user goals and objectives to underlying model(s) using VA systems. In the end, I summarize my contribution(s) and then reflect on assumptions, strengths, and limitations of the research to inform potential future direction of this work.

 

MS Teams Meeting InvitationMedium: (Via MS Teams on Chrome/MS Edge browser or MS Teams App.) 

Medium: (Via MS Teams on Chrome/MS Edge browser or MS Teams App.) 

________________________________________________________________________________

Microsoft Teams meeting

Join on your computer or mobile app

Click here to join the meeting

Learn More | Meeting options

Status

  • Workflow Status:Published
  • Created By:Tatianna Richardson
  • Created:11/12/2020
  • Modified By:Tatianna Richardson
  • Modified:11/12/2020

Categories

Keywords