cs 730 830 intro ai
play

CS 730/830: Intro AI Class Outro AI at UNH Wheeler Ruml (UNH) - PowerPoint PPT Presentation

CS 730/830: Intro AI Class Outro AI at UNH Wheeler Ruml (UNH) Lecture 27, CS 730 1 / 12 Class Outro The AI View Past Present Talk Paper Future Evaluations AI at UNH Class Outro Wheeler Ruml (UNH) Lecture 27,


  1. CS 730/830: Intro AI Class Outro AI at UNH Wheeler Ruml (UNH) Lecture 27, CS 730 – 1 / 12

  2. Class Outro ■ The AI View ■ Past ■ Present ■ Talk ■ Paper ■ Future ■ Evaluations AI at UNH Class Outro Wheeler Ruml (UNH) Lecture 27, CS 730 – 2 / 12

  3. The AI View of An Agent Class Outro ■ The AI View ■ Past ■ Present ■ Talk ■ Paper ■ Future ■ Evaluations AI at UNH percepts → → actions Wheeler Ruml (UNH) Lecture 27, CS 730 – 3 / 12

  4. Past perception : supervising learning (handwriting recognition), ■ Class Outro unsupervised learning (shape finding) ■ The AI View ■ Past [ HMMs ] ■ Present ■ Talk reasoning : constraint satisfaction, propositional satisfiability, ■ ■ Paper first-order logic theorem proving ■ Future ■ Evaluations [ tree search, optimization ] AI at UNH planning : state-space search, motion planning, ■ domain-independent task planning, planning under uncertainty (MDPs) [ anytime and real-time planning, reinforcement learning ] acting : filtering (MCL) ■ [ control ] Not: cognitive modeling, ethics, NLP, vision, philosophy of mind Wheeler Ruml (UNH) Lecture 27, CS 730 – 4 / 12

  5. Present Fri May 1: no recitation ■ Class Outro Tue May 5 9-noon: project presentations ■ The AI View ■ ■ Past 10+2 minutes/person ■ Present ■ Talk Mon May 11 2pm: final papers ■ ■ Paper email PDF, tarball, HOWTO ■ Future ■ Evaluations given tarball and HOWTO, raw results should be AI at UNH reproducible on agate Wheeler Ruml (UNH) Lecture 27, CS 730 – 5 / 12

  6. Tips for A Research Talk problem (example!), approach, results, extensions ■ Class Outro practice beforehand: word choice, timing ■ The AI View ■ ■ Past ■ Present ■ Talk ■ Paper ■ Future ■ Evaluations AI at UNH Wheeler Ruml (UNH) Lecture 27, CS 730 – 6 / 12

  7. Tips for a Research Paper use the standard form: introduction (motivate and define ■ Class Outro problem, summarize paper), previous work, your approach, ■ The AI View ■ Past experimental results, discussion, conclusion ■ Present ■ Talk write for someone who has taken an AI class but doesn’t ■ ■ Paper know anything about your specific problem ■ Future ■ Evaluations don’t just plot results, explicitly describe what they show and ■ AI at UNH the conclusions you draw from them Wheeler Ruml (UNH) Lecture 27, CS 730 – 7 / 12

  8. Future UNH AI group: usually weekly (Google ‘UNH AI group’) ■ Class Outro sign up for the mailing list! ■ The AI View ■ Past ■ Present Fall: ■ Talk ■ Paper ( Wheeler Ruml: CS 931 Planning for Robots ) ■ ■ Future ■ Evaluations ( Momotaz Begum: CS 733/833 Mobile Robotics ) ■ AI at UNH ( Laura Dietz: CS 753/853 Information Retrieval ) ■ Laura Dietz: CS 780/880 ML for Sequences and Text ■ Marek Petrik: CS 950 Reinforcement Learning ■ Spring: Marek Petrik: CS 750/850 Machine Learning ■ Marek Petrik and Mark Lyon: CS 757/857 Optimization ■ Momotaz Begum: CS 780/880 Computer Vision ■ Momotaz Begum: CS 933 Human-Robot Interaction ■ Laura Dietz: CS 953 Knowledge Graphs and Text ■ Wheeler Ruml (UNH) Lecture 27, CS 730 – 8 / 12

  9. Evaluations These are important! I take them seriously and so does my boss. Class Outro ■ The AI View ■ Past For free response text, please address: ■ Present ■ Talk 1. Things that were good about the class , things that need ■ Paper ■ Future work. ■ Evaluations specific suggestions or general comments! AI at UNH 2. Things that I did well, things that I should work on. Things that Tianyi did well, things that Tianyi should work on Thanks. Wheeler Ruml (UNH) Lecture 27, CS 730 – 9 / 12

  10. Class Outro AI at UNH ■ AI at UNH ■ EOLQs AI at UNH Wheeler Ruml (UNH) Lecture 27, CS 730 – 10 / 12

  11. AI at UNH Marek Petrik: robust RL ■ Class Outro Momotaz Begum: assistive robotics ■ AI at UNH ■ AI at UNH Laura Dietz: Queripedia ■ ■ EOLQs Wheeler Ruml: heuristic search, planning ■ rational real-time search (Tianyi) ◆ suboptimal and bounded suboptimal (William) ◆ real-time path coverage (Alex) ◆ online goal recognition design (Kevin) ◆ motion planning in a dynamic environment (Yi) ◆ group assignment (Brendan) ◆ ROP attack assembly (Daroc) ◆ physical TSP (Bryan, Lucas, Charles) ◆ situated temporal planning (Shahaf) ◆ Wheeler Ruml (UNH) Lecture 27, CS 730 – 11 / 12

  12. EOLQs Nope. Class Outro AI at UNH ■ AI at UNH ■ EOLQs Wheeler Ruml (UNH) Lecture 27, CS 730 – 12 / 12

Recommend


More recommend