~/talk

How the Experts Do It: Assessing and Explaining Agent Behaviors in Real-Time Strategy Games

on YouTube 18m
Talkat CHI 2018 · 2018-04-21..04-26· [machine-learning]

speakers

name:Jonathan Dodge
links:none on file
name:Sean Penney
links:none on file
name:Claudia Hilderbrand
links:none on file
name:Andrew Anderson
links:none on file
name:Margaret Burnett
links:none on file

resources

video:https://youtu.be/j2NqIW-Suu8
slides:none on file

summary

Collected and coded videos of professional shoutcasters and pro matches, then classify where they look using the PEAS framework into Performance Environment Actuators Sensors and model their navigation as information foraging. Key findings: the production tab (an actuator view) stays always on, performance metrics appear only about one per match, and explainers emphasize approximate why via satisficing rather than deep causal inference. Recommendations: expose actuator and sensor views and surface common abstractions for natural language generation.

AI-written summary · about

chapters (11)

  1. 0:00Introduction and motivation
  2. 1:49Shoutcasters as expert explainers
  3. 3:13Theoretical framework and data collection
  4. 4:58PIECE framework for agent behavior
  5. 6:43Frequently used information patches
  6. 8:06Navigation between information patches
  7. 9:51Types of questions answered by shoutcasters
  8. 11:58Why explanations are challenging
  9. 13:44Properties and patterns in explanations
  10. 15:29Implications and conclusions
  11. 16:11Q&A session

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.