comp 150 developmental robotics
play

COMP 150: Developmental Robotics Instructor: Jivko Sinapov - PowerPoint PPT Presentation

COMP 150: Developmental Robotics Instructor: Jivko Sinapov www.cs.tufts.edu/~jsinapov Language Acquisition Something fun... Announcements Project Deadlines Project Presentations: Dec 5 and 7 Final Report + Deliverables: Dec 11


  1. COMP 150: Developmental Robotics Instructor: Jivko Sinapov www.cs.tufts.edu/~jsinapov

  2. Language Acquisition

  3. Something fun...

  4. Announcements

  5. Project Deadlines ● Project Presentations: Dec 5 and 7 ● Final Report + Deliverables: Dec 11 ● Deliverables: – Presentation slides + videos – Final Report (PDF) – Source code (link to github repositories)

  6. Presentation Guidelines ● 10 minutes talk + 5 min for questions ● Practice! Time your presentation when you practice and use a timer during the actual presentation as well ● My advice: find another group and practice to each other ● Format: Google Slides (so that we don’t have to switch computers)

  7. Presentation Schedule – Tue Dec 5 ● Raina Galbiati, Doo-yun Her, and Cassie Collins ● Azmina Karukappadath, Sam Weiss, and Yuelin Liu ● Timi Dayo-Kayode, Michael Edegware, and Jong Seo Yoon ● Matt Ryan ● Meghan O'Brien, Tooba Ahsen, and Elizabeth Lanzilla

  8. Presentation Schedule – Thu Dec 7 ● Julia Novakoff, Teddy Laurita, and George Pesmazoglou ● Eric Chen, Matt Shenton, and Avi Block ● Ari Brown and Julie Jiang ● Christopher Hylwa, Sonal Chatter, and Brett Gurman ● Brad Oosterveld, Tyler Frascav

  9. Final Report Guidelines ● Approximately 8 pages + 1 page for references ● Default Google Doc template or default overleaf LaTeX template ● A bit about the structure… ● May include appendix if you have a lot of visual results

  10. Last Readings and Homework ● See course website

  11. Language Acquisition

  12. The Turing Test

  13. The Turing Test

  14. The Turing Test

  15. The First ChatBot (~1966)

  16. ELIZA ● http://psych.fullerton.edu/mbirnbaum/psych101/ Eliza.htm

  17. Discussion: what is missing from programs like ELIZA?

  18. Natural Language Processing ● The study of algorithms and data structures used to manipulate text and text-like data ● Applications in information retrieval, web search, dialogue agents, text mining, etc. ● Traditionally, not concerned with connecting semantic representations to the real world

  19. Example: Computing Parse Trees

  20. Example: Document Classification https://abbyy.technology/_media/en:features:classification- scheme.png

  21. Example: Word Embeddings https://image.slidesharecdn.com/introductiontowordembeddings-160405062343/95/a-simple-introduction-to-word-embeddings-5-638.jpg?cb=1494520542

  22. The Symbol Grounding Problem “How can the semantic interpretation of a formal symbol system be made intrinsic to the system, rather than just parasitic on the meanings in our heads? How can the meanings of the meaningless symbol tokens, manipulated solely on the basis of their (arbitrary)shapes, be grounded in anything but other meaningless symbols?” - Steven Hamas, 1990

  23. Deb Roy, “Grounding Language in the World: Schema Theory Meets Semiotics” (2005)

  24. Circular Definitions

  25. Grounding

  26. Sensor Projections

  27. Sensor Projections INPUT IMAGE Color Histogram

  28. Transformer Projection

  29. Transformer Projection Color Histogram Entropy of Histogram

  30. Categorizer Entropy of Histogram “Multicolored”

  31. Action Projector

  32. Schemas for Actions

  33. Schemas for Objects

  34. Spatial Relations

  35. Deb Roy’s Definition of Grounding ● “I define grounding as a causal-predictive cycle by which an agent maintains beliefs about its world.” (p. 8) ● “An agent’s basic grounding cycle cannot require mediation by another agent.” (p. 9) ● “An autonomous robot simply cannot afford to have a human in the loop interpreting sensory data on its behalf.” (p. 9)

  36. ● “Cyclic interactions between robots and their environment, when well designed, enable a robot to learn, verify, and use world knowledge to pursue goals. I believe we should extend this design philosophy to the domain of language and intentional communication.” (p. 5)

  37. ● “causality alone is not a sufficient basis for grounding beliefs. Grounding also requires prediction of the future with respect to the agent’s own actions.” (p. 10) ● “The problem with ignoring the predictive part of the grounding cycle has sometimes been called the ”homunculus problem”.”

  38. Take Home Message Language should be grounded in terms of the robot’s own perceptual and sensorimotor capabilities

  39. Thomason, J., Sinapov, J., Svetlik, M., Stone, P., and Mooney, R. (2016) Learning Multi-Modal Grounded Linguistic Semantics by Playing I, Spy In proceedings of the 2016 International Joint Conference on Artificial Intelligence (IJCAI)

  40. Motivation: Grounded Language Learning Robot, fetch me the green empty bottle 43

  41. Exploratory Behaviors in our Robot 44

  42. Video 45

  43. Video 46

  44. Video 47

  45. Sensorimotor Feature Extraction . . . . . . Joint Efforts (Haptics) Time 48

  46. Sensorimotor Contexts proprio- haptics audio shape color VGG ception look grasp lift hold lower drop push press 49

  47. Sensorimotor Contexts proprio- haptics audio shape color VGG ception look grasp lift hold lower drop push press 50

  48. Feature Extraction: Color Object Segmentation Color Histogram (4 x 4 x 4 = 64 bins) 51

  49. Feature Extraction: Shape 3D Object Point Cloud Histogram of Shape Features 52

  50. Feature Extraction: Haptics Joint-Torque values for all joints Joint-Torque Features 53

  51. Feature Extraction: Audio audio spectrogram Spectro-temporal Features 54

  52. Feature Extraction: VGG 55

  53. Feature Extraction: VGG 56

  54. Data from a single exploratory trial proprio- haptics audio shape color VGG ception look grasp lift hold lower drop push press x 5 per object 57

  55. Category Recognition Overview Interaction with Object Category Estimates Red? . . . Container? Empty? Sensorimotor Feature Category Extraction Recognition Models Sinapov, J., Schenck, C., and Stoytchev, A. (2014). Learning Relational Object Categories Using Behavioral Exploration and Multimodal Perception In the Proceedings of the 2014 IEEE International Conference on Robotics and Automation (ICRA) 58

  56. Key Questions How can the robot learn object-related words from everyday human users? Do human users use non-visual object descriptors when referring to objects? 59

  57. Object Exploration Dataset 32 common household and office items Each object was explored a total of 5 times with 7 different behaviors The robot perceived objects using the visual, auditory, and haptic sensory modalities Thomason, J., Sinapov, J., Svetlik, M., Stone, P., and Mooney, R. (2016). Learning Multi-Modal Grounded Linguistic Semantics by Playing I, Spy In proceedings of the 2016 International Joint Conference on Artificial Intelligence (IJCAI) 60

  58. Our attempt: I-Spy game 61

  59. Learning Words via Game-play Human: “an empty metallic aluminum container” 62

  60. Semantic Parsing 63

  61. Example Words for an Object 64

  62. Learning Words via Game-play 65

  63. Learning Words via Game-play Human: “a tall blue cylindrical container” 66

  64. Learning Words via Game-play Robot: “open half-full container” 67

  65. Asking Verification Questions 68

  66. Results 69

  67. F-measure improvement as WORD a result of adding non- visual modalities “can” 0.857 “tall” 0.516 “half-full” 0.463 . . . . . . . . “pink” 0 70

  68. Summary of Experiment ● The robot learned over 80 words through interactive game play ● The robot's word representations were grounded in multiple behaviors and sensory modalities ● Future Work: – Active action selection when classifying a new object – Active action selection when learning a new words – Actively seek humans out for help with learning about objects 71

  69. “Opportunistic” Active Learning Thomason, J., Padmakumar, A., Sinapov, J., Hart, J., Stone, P., and Mooney, R. (2017) Opportunistic Active Learning for Grounding Natural Language Descriptions In proceedings of the 1st Annual Conference on Robot Learning (CoRL 2017) 72

  70. “Opportunistic” Active Learning Thomason, J., Padmakumar, A., Sinapov, J., Hart, J., Stone, P., and Mooney, R. (2017) Opportunistic Active Learning for Grounding Natural Language Descriptions In proceedings of the 1st Annual Conference on Robot Learning (CoRL 2017) 73

  71. What actions should the robot perform when learning a new word? ● Baseline: perform all actions on a set of labeled objects and estimate which ones work well ● But can we do better? 74

  72. Behavior Scores for Words 75

  73. Word Embeddings Thomason, J., Sinapov, J., Stone, P., and Mooney, R. (2018) Guiding Exploratory Behaviors for Multi-Modal Grounding of Linguistic Descriptions To appear in proceedings of the 32nd Conference of the Association for the Advancement of Artificial Intelligence (AAAI) 76

Recommend


More recommend