Artificial Intelligence

Transkript

1 Artificial Intelligence Introduction Marc Toussaint University of Stuttgart Winter 2015/16 (slides based on Stuart Russell s AI course)

2 The rise of AI (again) The singularity Ethics The value problem The outrageous inability of humans to define what is good Paper clips 2/13

3 What s the route to AI? Neuroscience? (EU Big Brain project) Deep Learning? (Pure Machine Learning?, DeepMind (London)) Social/Emotional/conciousnes/Feelings stuff? Hardcore classical AI? Modern probabilistic/learning AI? Robotics? Why is there no university department for Intelligence Research?! 3/13

4 Potted history of AI 1943 McCulloch & Pitts: Boolean circuit model of brain 1950 Turing s Computing Machinery and Intelligence Look, Ma, no hands! 1950s Early AI programs, including Samuel s checkers program, Newell & Simon s Logic Theorist, Gelernter s Geometry Engine 1956 Dartmouth meeting: Artificial Intelligence adopted 1965 Robinson s complete algorithm for logical reasoning AI discovers computational complexity Neural network research almost disappears Early development of knowledge-based systems Expert systems industry booms Expert systems industry busts: AI Winter Neural networks return to popularity 1988 Resurgence of probability; general increase in technical depth Nouvelle AI : ALife, GAs, soft computing 1995 Agents, agents, everywhere Human-level AI back on the agenda 4/13

5 What is intelligence? Maybe it is easier to first ask what systems we actually talk about: Decision making Problem solving Interacting with an environment Then define objectives 5/13

6 Interactive domains We assume the agent is in interaction with a domain. The world is in a state s t S (see below on what that means) The agent senses observations y t O The agent decides on an action a t A The world transitions to a new state s t+1 The observation y t describes all information received by the agent (sensors, also rewards, feedback, etc) if not explicitly stated otherwise (The technical term for this is a POMDP) 6/13

7 Worlds do not have additional latent (hidden) variables to the state s t 7/13 State The notion of state is often used imprecisely At any time t, we assume the world is in a state s t S s t is a state description of a domain iff future observations y t +, t + > t are conditionally independent of all history observations y t, t < t given s t and future actions a t:t +: agent y 0 a 0 y 1 a 1 y 2 a 2 y 3 a 3 s 0 s 1 s 2 s 3 Notes: Intuitively, s t describes everything about the world that is relevant

8 Examples What is a sufficient definition of state of a computer that you interact with? What is a sufficient definition of state for a thermostat scenario? (First, assume the room is an isolated chamber.) What is a sufficient definition of state in an autonomous car case? 8/13

9 Examples What is a sufficient definition of state of a computer that you interact with? What is a sufficient definition of state for a thermostat scenario? (First, assume the room is an isolated chamber.) What is a sufficient definition of state in an autonomous car case? in real worlds, the exact state is practically not representable all models of domains will have to make approximating assumptions (e.g., about independencies) 8/13

10 How can agents be formally described?...or, what formal classes of agents do exist? Basic alternative agent models: The agent maps y t a t (stimulus-response mapping.. non-optimal) The agent stores all previous observations and maps f : y 0:t, a 0:t-1 a t f is called agent function. This is the most general model, including the others as special cases. The agent stores only the recent history and maps y t k:t, a t k:t-1 a t (crude, but may be a good heuristic) The agent is some machine with its own internal state n t, e.g., a computer, a finite state machine, a brain... The agent maps (n t-1, y t) n t (internal state update) and n t a t The agent maintains a full probability distribution (belief) b t(s t) over the state, maps (b t-1, y t) b t (Bayesian belief update), and b t a t 9/13

11 POMDP coupled to a state machine agent agent n 0 n 1 n 2 y 0 a 0 y 1 a 1 y 2 a 2 s 0 s 1 s 2 r 0 r 1 r 2 Is this a very limiting agent model? 10/13

12 Multi-agent domain models (The technical term for this is a Decentralized POMDPs) (from Kumar et al., IJCAI 2011) This is a special type (simplification) of a general DEC-POMDP Generally, this level of description is very general, but NEXP-hard Approximate methods can yield very good results, though 11/13

13 We gave above very general and powerful models (formalizations) of what it means that an agent takes decisions in an interactive environment. There are many flavors of this: Fully observable vs. partially observable Single agent vs. multiagent Deterministic vs. stochastic Structure of the state space: Discrete, continuous, hybrid; factored; relational Discrete vs. continuous time Next we need to decide on Objectives 12/13

14 Objectives, optimal agents, & rationality Utilities, rewards, etc An optimal (=rational) agent is one that takes decisions to maximize the (expected) objective 13/13

15 sequential decisions propositional relational deterministic sequential decision problems search BFS sequential assignment games backtracking alpha/beta pruning minimax CSP constraint propagation on trees propositional logic fwd/bwd chaining FOL MCTS UCB bandits learning probabilistic Decision Theory utilities multi-agent MDPs Reinforcement Learning MDPs dynamic programming V(s), Q(s,a) relational MDPs Active Learning graphical models ML belief propagation msg. passing HMMs FOL fwd/bwd msg. passing relational graphical models 14/13

16 Organisation 15/13

17 Vorlesungen der Abteilung MLR Bachelor: Grundlagen der Künstlichen Intelligenz (3+1 SWS) Master: Vertiefungslinie Intelligente Systeme (gemeinsam mit Andres Bruhn) WS: Maths for Intelligent Systems WS: Introduction to Robotics SS: Machine Learning SS: Optimization manchmal: Reinforcement Learning (Vien Ngo), Advanced Robotics Hauptseminare: Machine Learning (WS), Robotics (SS) 16/13

18 Andres Bruhn s Vorlesungen in der Vertiefungslinie WS: Computer Vision SS: Correspondence Problems in Computer Vision Hauptseminar: Recent Advances in Computer Vision 17/13

19 Vorraussetzungen für die KI Vorlesung Mathematik für Informatiker und Softwaretechniker außerdem hilfreich: Algorithmen und Datenstrukturen Theoretische Informatik 18/13

20 Vorlesungsmaterial Webseite zur Vorlesung: die Folien und Übungsaufgaben werden dort online gestellt Alle Materialien des letzten Jahres sind online bitte machen Sie sich einen Eindruck Hauptliteratur: Stuart Russell & Peter Norvig: Artificial Intelligence A Modern Approach Many slides are adopted from Stuart 19/13

21 Prüfung Schriftliche Prüfung, 60 Minuten Termin zentral organisiert keine Hilfsmittel erlaubt Anmeldung: Im LSF / beim Prüfungsamt Prüfungszulassung: 50% der Punkte der Übungsaufgaben 20/13

22 Übungen 8 Übungsgruppen (4 Tutoren) 2 Arten von Aufgaben: Coding- und Votier-Übungen Coding-Aufgaben: Teams von bis zu 3 Studenten geben die Coding-Aufgaben zusammen ab Votier-Aufgaben: Zu Beginn der Übung eintragen, welche Aufgaben bearbeiten wurden/präsentiert werden können Zufällige Auswahl Schein-Kriterium: 50% aller Aufgaben gelöst (Coding und Voting). Registrierung username: KI, passwd: ALS VORNAMEN ANGEBEN: Vorname TEAMNAME, zb Peter frogs 21/13