BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:PhD Prelim Exam - Jacob Krantz
X-WR-TIMEZONE:Pacific Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260719T095625Z
UID:tag:localist.com\,2008:EventInstance_43349475434818
DTSTART:20230608T230000Z
DTEND:20230609T010000Z
DESCRIPTION:Semantic Embodied Navigation: Developing Agents That Navigate F
 rom Language and Vision\n\nAutonomous robotic agents are on their way to b
 ecoming in-home personal assistants\, construction assistants\, and wareho
 use workers. The degree of autonomy of such systems is reflected by the ma
 nner in which we specify our goals to them\; the abstraction of low-level 
 commands to high-level goals goes hand-in-hand with increased autonomy. In
  this work\, we are interested in developing artificial intelligence that 
 enables agents to accomplish semantic goals. We specifically study semanti
 c embodied navigation. In embodied navigation\, agents act in spatial envi
 ronments to follow a motion path or reach a destination. The qualifier sem
 antic refers to how the goal is specified\; instead of directly (e.g. coor
 dinates)\, a goal is provided in a manner that either implies the goal or 
 constrains the possible interpretations of the goal. We study two abilitie
 s of semantic embodied navigation: navigating in response to natural langu
 age instructions and navigating to an object specified by an image. Our co
 ntributions are aimed at developing fundamental semantic-guided navigation
  abilities for embodied agents and are highlighted as follows. For both ta
 sks\, we establish benchmarks in simulation and evaluate baseline models. 
 For language-based navigation\, we develop language-conditioned waypoint p
 rediction networks to study the impact of mid-level action spaces. We then
  show that by transferring an agent from highly-abstract simulation to low
 er-level simulation\, we can achieve high performance on our language navi
 gation benchmark. For image-based navigation\, we find that existing metho
 ds that train sensors-to-action policies struggle to generalize to new env
 ironments and new goals. We propose decomposing the image navigation task 
 into sub-tasks addressable via modular components. We will evaluate the re
 sulting agent both on our simulation benchmark and demonstrate performance
  in the real world.\n\nMAJOR ADVISOR: Stefan Lee\nCOMMITTEE: Prasad Tadepa
 lli\nCOMMITTEE: Alan Fern\nCOMMITTEE: Fuxin Li\nGCR: Joseph Louis
GEO:44.567164;-123.278692
LOCATION:Kelley Engineering Center\, 2057
SUMMARY:PhD Prelim Exam - Jacob Krantz
URL;VALUE=URI:https://events.oregonstate.edu/event/phd_prelim_exam_-_jacob_
 krantz
CATEGORIES:Lecture or Presentation
END:VEVENT
END:VCALENDAR