Conversation as Action Under Uncertainty

1
Conversation as Action Under Uncertainty

• Tim Paek
• Eric Horvitz

2
Abstract

• Conversation
• uncertainties of various kinds
• Quartet architecture
• Inference/Decision making under uncertainty
• 4 levels of analysis
• Representations
• Inference procedures
• Decision strategies
• Interaction between a user and dialog system

3
1. Introduction

• Spoken dialog system
• robust, unified architecture for sharing key
  uncertainties ? Quartet

4
2. Toward an Architecture for Conversation

• General conversation
• To make themselves understood
• Uncertainty communication failure
• Collaborate/establish mutual belief
• feedback coordination
• Head nod, uh huh
• Clear up uncertainties, take other measures
• Conversation collaborative effort or joint
  activity

5

• Grounding
• ?? ??? ?? ??? presentation ?? acceptance?
  coordinate ?? ??
• Consideration of how key uncertainties depend on
  each other and influence mutual understanding
• What decisions to make in light of these
  uncertainties

6

• Quartet
• Treat grounding as decision making under
  uncertainty
• Key uncertainty ? Bayesian Network
• Local expected utility/VOI
• Identify action ? maximize mutual understanding
• Focus on handling uncertainties
• Enable dialog system to adapt strategies
• Allow components to share uncertainties

7
2.1 Four Levels of Analysis

• Evaluate grouding at 4 levels of analysis
• 4 Levels
• Channel level
• Signal level
• Intention level signal? semantics ??
• Conversation level
• All levels need coordination/collaboration
• Unified architecture
• Model probabilistic dependencies between levels

8
(No Transcript)
9
2.2 Models, Inference, Decisions

• atemporal/temporal Bayesian networks
• compute likelihood of variables of interest
• Ex) overall grounding status

10
2.2.1 Modularity

• Maintenance module
• handle uncertainties about channel and signal
  levels
• Intention module
• About intention level
• Conversation control
• Keep track of the grounding status
• Where to focus on resolving uncertainties
• What grounding actions to take in light of their
  likely costs/benefits

11

• Reasons for distinct modules
• Probabilistic dependencies
• abundant with the modules
• sparse between them
• Enhancement of the flexibility of Quartet
• To be applied to variety of task domains
• Belief, decision, degree of detailed necessary
  for each level varied significantly depending on
  the communication medium or modality

12
(No Transcript)
13
(No Transcript)
14
2.2.2 Representation for Decision Making

• Grouding status
• Diagnoses the overall mutual understanding

15
2.2.3 Refinement through Value of Information

• Need to refine actions by sharing uncertainties
• Utterance design
• improve understanding
• Conversation ? decision making
• Capitalize on VOI analysis to ask questions, make
  recommendations, and seek out information

16

• VOI analysis
• find best evidence in light of the inferred
  probabilities
• Compute VOI
• ?? observation? ?? expected utility of the best
  decision? ??
• VOI recommend a piece of evidence to observe
  system? recommendation ?? ?? phrase a question

17

• Exact computation of VOI ? all possible sequences
  of observations
• Greedy VOI reasonable approximation

18
3. Runtime Behavior

• Dialog system
• ASR
• NLPWin parser
• Face-pose tracker

19
3.1 Providing Services

• ?? ???? highest expected utility action
• User? requested action? ????? ?

20
(No Transcript)
21
3.2 Exploiting Multiple Levels

• Goal? not so clear at the intention level
• Consider uncertainties at lower levels of analysis

22
(No Transcript)
23
3.3 Adapting Strategies over Time

• Mutual understanding? fluctuate ? adapt strategy
  over time
• Do requested service / resort to only three
  repair strategies
• ASK REPEAT, CONFIRM, TROUBLESHOOT (use VOI
  analyses to recommend procedures to improve
  mutual understanding)

24
(No Transcript)
25
4. Conclusions

• Approach to continuous spoken dialog centering on
  an architecture called Quartet