How to AI COGS 105 Many robotics and engineering problems work from - PowerPoint PPT Presentation

How to AI COGS 105 • Many robotics and engineering problems work from a task- Week 14b: AI and Robotics based perspective (see competing traditions from last class). • What is your task ? What are the inputs and outputs to your agent ? • What are the rules that lead from input to output? Example if-then rule in JavaScript: Traditional Rule Concept == compares what 
 user said (wus) to “hi” • In many respects, classic AI started with the concept of a production rule . if (wus=="hi") { was = "hi back to ya"; • An if-then rule that determines what the agent does if some } condition holds. • Production rules can be more complicated by using pattern- Simon & Newell, 50’s matching (and pattern-generating) = sets what it says (was) to “hi back...” algorithms .

Other pattern matches... Where Do 
 Rules Come From? indexOf = check to see if the word “mother” appears in what they said. • Imagine trying to build an artificial agent that can engage in conversations . if (wus.indexOf("mother")>=0) { ... • What does the agent need to know? if (wus=="hi" || wus=="hello" || wus=="hey" || wus=="hiya") { ... • Words if (wus.length > 10) { ... • How to line up words into sentences • How to choose which sentence topics to pursue || = “or”, to check length = how long, in • How to stay on topic multiple possible patterns characters, is what they said? • How to guide the flow of conversation (e.g., question answer) Where Do 
 Where Do 
 Rules Come From? Rules Come From? • Imagine trying to build an artificial agent that can engage in conversations . • What does the agent need to know? • Morphology • Syntax • Semantics • Pragmatics / Discourse • Pragmatics / Discourse 
 http://talkbank.org/

Conversation Analysis “SEMAINE” Project • The analysis of interaction using video and audio, and then careful hand-coding and observation. • Qualitative method. • Many of the methods in conversation analysis can be transformed, in simple ways, into production rules. For example, the adjacency pair . • E.g., offer/refusal (or acceptance), question- answer, compliment-response Another: SimCoach Example: USC ICT Projects • USC’s Institute for Creative Technologies • They develop sophisticated intelligent virtual characters that can engage in interaction using sets of sophisticated rules. http://prod1.standardpatient.org/

Cutting Edge: 
 ELITE Authoring Tools • The field of virtual agents is especially active in developing and testing authoring tools , to make it easier for researchers or people in industry to create artificial agents. • These tools have a “point and click” strategy, and allow users to build large repositories of pattern- matching algorithms (production rules, with some spice) without having to program. RoundTable • https://authoring.simcoach.org/

demo from previous year demo from previous year (just for illustration) (just for illustration) A Simple Interpreter Designed for this Purpose demo from previous year (just for illustration)

Two AI Cultures cognaction.org/ Embodied cogs105/chatagent GOFAI-based… Dynamic Situated Hybrid Systems… Robotics / AI Various Issues in AI, Etc. • The field is massive. Even just the tools built into this artificial, virtual software agent exemplify the • One problem faced by array of domains of this field. engineers and scientists is that robotic systems are becoming action modeling complicated enough that HCI it is difficult to do “on speech recognition board.” speech synthesis • Solution: The cloud. computational linguistics From reading #2

Benefits Involvement of CogSci • Increased computational power “off board”: using • If you know how to code , you can be involved in all high-performance servers to process / crunch stages… numbers off the system itself. • …machine learning, neural networks, computer • Increased distribution of robotic (or AI) agents vision, etc. can solve a problem faster, “in parallel.” • If you do not or do not want to code , you can still be involved by… • …designing usability studies, serving as a “domain expert” to help coders, analyze human behavior, etc. From reading #2 Phamduy et al. Project Ideas • I can open up the chatbot system for you to create your own chat agent. • You can use the Braitenberg system to do some exploration of agent-based simulations (extending the last lab). • You can do a usability study (ask me for readings, explaining how to do it). with children