1 / 23

Asking Questions and Developing Trust

Asking Questions and Developing Trust. Stephanie Rosenthal Joint Work with Anind K. Dey and Manuela Veloso Carnegie Mellon University. Overview. Questions. Agent/ Robot. Human(s). Responses. Agents Asking Questions. Agent should explicitly share its state information

viveka
Download Presentation

Asking Questions and Developing Trust

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Asking Questions and Developing Trust Stephanie Rosenthal Joint Work with Anind K. Dey and Manuela Veloso Carnegie Mellon University

  2. Overview Questions Agent/ Robot Human(s) Responses Asking Questions and Developing Trust

  3. Agents Asking Questions • Agent should explicitly share its state information • context, prediction, uncertainty, features • Goal: Vary which information the agent shares to maximize accuracy Robot Email Sorter Asking Questions and Developing Trust

  4. Questions - Robot Example • Robot State • Context - image of the structure, color, position • Prediction - shape of block • Uncertainty - probability the prediction is wrong • Features - context that defines the block Asking Questions and Developing Trust

  5. Questions - Robot Example • Robot State • Context - image of the structure, color, position • Prediction - shape of block • Uncertainty - prob. prediction is wrong • Features - context that defines the block • Example Question • Cannot determine the block shape. You are working with the red and purple blocks. What shape is the red block? I think it is a triangular prism. Asking Questions and Developing Trust

  6. Questions - Robot Example • Robot State • Context - image of the structure, color, position • Prediction - shape of block • Uncertainty - prob. prediction is wrong • Features - context that defines the block • Example Questions • You are working with the red and purple blocks. What shape is the red block? What features define that shape? Asking Questions and Developing Trust

  7. Findings Robot Email Asking Questions and Developing Trust

  8. Overview Agents can manipulate their questions to maximize the accuracy of human responses Questions Agent/ Robot Human(s) Responses Asking Questions and Developing Trust

  9. Recommender System Goal: Provide personal predictions for each user N products p1 p2 … pN R reviews r1 r2 … rM Rij M reviewers c1 c2 cK Categories Asking Questions and Developing Trust

  10. Domain-Independent Initialize wiu = 1/M For all products pj that user u requests predictions for: Make prediction argmaxv I(Rij==v)wu If user gives opinion oj Update weights wu Domain-Specific Initialize wiu,d = 1/M For all products pj that user u requests predictions for: d = domain of pj Make prediction argmaxv I(Rij==v)wu,d If user gives opinion oj Update weights wu,d Advice Givers Good: Reduces sparsity Bad: Single set of weights Good: Category-based weights Bad: Less data in each category Asking Questions and Developing Trust

  11. Which Advice Giver is Better? • Tradeoff between data and precision is not uniform across users • User-dependent selection algorithm to decide which advice giver is best for each user DI DS Selection Asking Questions and Developing Trust

  12. Summary Agents can manipulate their questions to maximize the accuracy of human responses Questions Agent/ Robot Human(s) Genre and frequency of questions affects the way that the agent should develop trust with reviewers Responses Asking Questions and Developing Trust

  13. Future Work Agents can manipulate their questions to maximize the accuracy of human responses Questions Agent/ Robot Human(s) Genre and frequency of questions affects the way that the agent should develop trust with reviewers Responses Asking Questions and Developing Trust *with Mike Licitra, Nick Armstrong-Crews, Joydeep Biswas

  14. Questions? Asking Questions and Developing Trust

  15. Asking Questions and Developing Trust

  16. Recommender System • Advice giver weighs each reviewer for each user wiu • For all users, initialize wiu = 1/M • When user u provides an actual opinion oj about a product pj, update all weights wu wiu = e^(ln(wiu) - |Rij - oj|)/K • Advice giver predicts value v for product pj and user u • argmaxv I(Rij==v)wu Asking Questions and Developing Trust

  17. Advice Giver Algorithm • Initialize wiu = 1/M • For all products pj that user u requests predictions for: • Make prediction argmaxv I(Rij==v)wu • If user gives opinion oj • Update weights wu Asking Questions and Developing Trust

  18. Category-Dependent Advice Giver Algorithm • Initialize wiu,k = 1/M • For all products pj that user u requests predictions for: • k = category of pj • Make prediction argmaxv I(Rij==v)wu,k • If user gives opinion oj • Update weights wu,k only Asking Questions and Developing Trust

  19. Tradeoffs • Category-Independent Advice Giver • More data to evaluate the weights of each reviewer, coarser trust model • Category-Dependent Advice Giver • More fine-grained evaluation of which reviewers to trust, less data per category • Amazon.com, Netflix.com, Yahoo! Music • V = {1,2,3,4,5}, M > 100K, N > 50 per user • C = ~10 per dataset, 20 test users per dataset Asking Questions and Developing Trust

  20. Overview • Asking Questions of Novice Users • Developing Trust in Large Sets of Online Users • CoBot the Visitor Companion Robot Asking Questions and Developing Trust

  21. Asking Questions • Agent should explicitly share its state information • context, prediction, uncertainty, features • Goal: Vary which information the agent shares to maximize accuracy Asking Questions and Developing Trust

  22. Developing Trust in Humans • Case Study - Recommender Systems Asking Questions and Developing Trust

  23. CoBot, Visitor Companion Escort a human visitor to their meetings • Navigate indoor environments • Share information relevant to the meetings • Ask questions when it cannot perform a task Asking Questions and Developing Trust

More Related