1CS 188: Artificial IntelligenceFall 2010Advanced Applications:Robotics / Vision / LanguageDan Klein – UC BerkeleyMany slides from Pieter Abbeel, John DeNero1Announcements Project 5: Classification up now! Due date now after contest Also: drop-the-lowest Contest: In progress! New staff bot (w/ extra credit) New achievements2So Far: Foundational Methods3Now: Advanced Applications4Web Search / IR Information retrieval: Given information needs, produce information Includes, e.g. web search, question answering, and classic IR Web search: not exactly classification, but rather rankingx = “Apple Computers”Feature-Based Rankingx = “Apple Computers”x,x,2Perceptron for Ranking Inputs Candidates Many feature vectors: One weight vector: Prediction: Update (if wrong):Inverse RL: Motivation How do we specify a task like this?[demo: hover / autorotate]Autonomous Helicopter SetupOn-board inertial measurement unit (IMU)Send out controls to helicopterPositionHelicopter MDP State: Actions (control inputs): alon: Main rotor longitudinal cyclic pitch control (affects pitch rate) alat: Main rotor latitudinal cyclic pitch control (affects roll rate) acoll: Main rotor collective pitch (affects main rotor thrust) arud: Tail rotor collective pitch (affects tail rotor thrust) Transitions (dynamics): st+1= f (st, at) + wt[f encodes helicopter dynamics][w is a probabilistic noise model] Can we solve the MDP yet?Problem: What’s the Reward? Rewards for hovering: Rewards for “Tic-Toc”? Problem: what’s the target trajectory? Just write it down by hand?11[demo: hover / tic-toc][demo: bad]Apprenticeship Learning Goal: learn reward function from expert demonstration Assume Get expert demonstrations Guess initial policy Repeat: Find w which make the expert better than Solve MDP for new weights w:123Pacman Apprenticeship! Demonstrations are expert games Features defined over states s Score of a state given by: Learning goal: find weights which explain expert actions[demo: pac apprentice]Helicopter Apprenticeship?14[demo: unaligned / aligned]Probabilistic Alignment Intended trajectory satisfies dynamics. Expert trajectory is a noisy observation of one of the hidden states. But we don’t know exactly which one.Intended trajectoryExpert demonstrationsTime indicesAlignment of Samples Result: inferred sequence is much cleaner!16[demo: alignment]Final Behavior17[demo: airshow]What is NLP? Fundamental goal: analyze and process human language, broadly, robustly, accurately… End systems that we want to build: Ambitious: speech recognition, machine translation, information extraction, dialog interfaces, question answering… Modest: spelling correction, text categorization…184Problem: Ambiguities Headlines: Enraged Cow Injures Farmer With Ax Hospitals Are Sued by 7 Foot Doctors Ban on Nude Dancing on Governor’s Desk Iraqi Head Seeks Arms Local HS Dropouts Cut in Half Juvenile Court to Try Shooting Defendant Stolen Painting Found by Tree Kids Make Nutritious Snacks Why are these funny?Parsing as Search20Grammar: PCFGs Natural language grammars are very ambiguous! PCFGs are a formal probabilistic model of trees Each “rule” has a conditional probability (like an HMM) Tree’s probability is the product of all rules used Parsing: Given a sentence, find the best tree – search!ROOT → S 375/420S → NP VP . 320/392NP → PRP 127/539VP → VBD ADJP 32/401…..21Syntactic AnalysisHurricane Emily howled toward Mexico 's Caribbean coast on Sunday packing 135 mph winds and torrential rain and causing panic in Cancun, where frightened tourists squeezed into musty shelters .22[demo]Machine Translation Translate text from one language to another Recombines fragments of example translations Challenges: What fragments? [learning to translate] How to make efficient? [fast translation search]24528Levels of TransferMachine Translation632[demo:
View Full Document
Unlocking...