@@What is Fuzzy Logic Fuzzy Logic was developed by Lotfi Zadeh at UC Berkley
Fuzzy logi is a branch of logic specially designed for representing knowledge
and human reasoning in such a way that it is amenable to processing by a oputer WHAT IS FUZZY LOGIC? Definition of fuzzy : Fuzzy – ot lear, distit, or preise; lurred Definition of fuzzy logic*A form of knowledge representation suitable for notions that cannot be defined precisely, but which depend upon their contexts.*FOUNDATION OF FUZZY SYSTEMS • Fuzzyness pertains to the uncertainty associated with systems. Nothing can be predicted with exact precision – Vagueness/uncertainty cannot be adequately expessed by crisp variable – So we need fuzzy variables and fuzzy functions Application using fuzzy logic** Power system stability control**Power system stability assessment** Line fault detection and optimization process for generation, transmission and distribution**Speed control**Wind turbine control**Motor efficiency optimization**Waveform estimation etc.. Advantages of using fuzzy logic** Fuzzy logic controllers are not dependent on accurate mathematical model**They are based on heuristics, and they are able to incorporate human intuition and experience From Crisp to fuzzy set: Crisp sets is the collection of distinct (precisely defined) elements.**In a crisp set, an element is either a member of the set or not.**For example, a jelly bean belongs in the classof food known as candy (super set). Mashed potatoes do not. @@Fuzzy sets,**allow elements to be partially in a set.** Each element is given a degree of membership in a set.**This membership value can range from 0 (not an element of the set) to 1 (a member of the set).**membership function is the relationship between the values of an element and its degree of membership in a set. • An example of membership functions are shown in Figure. In this example, the sets (or classes) are numbers that are negative large, negative medium, negative small, near zero, positive small, positive medium, and positive large. The value, μ, is the amount of membership in the set. @@Expert System An expert system is a system that employs human knowledge captured in a computer to solve problems that ordinarily require human expertise.(Turban) A computer program that emulates the behaviour of human experts who are solving real-world problems associated with a particular domain of knowledge. (Pigford & Braur)@@Expert System Expert Systems manipulate knowledge while conventional programs manipulate data. An expert system is often defined by its structure. Knowledge Based System Vs Expert System.@@Characteristics of an Expert System:**Pigford & Baur Inferential Processes **Uses various Reasoning Techniques Heuristics **Decisions based on experience and knowledge .@@Knowledge and Uncertainty Facts and rules are structured into a knowledge base and used by expert systems to draw conclusions. There is often a degree of uncertainty in the knowledge. **Things are not always true or false **the knowledge may not be complete. In an expert system certainty factors are one way indicate degree of certainty attached to a fact or rule. @@Classification of Expert Systems:**Classification based on “Expertness” or Purpose **Expertness @@An assistant **used for routine analysis and points out those portions of the work where the human expertise is required. **A colleague the user talks over the problem with the system until a “joint decision” is reached. **A true expert the user accepts the system’s advice without question. @@Components of an Expert System:daiagram @@Desirable Features of an Expert System :Dealing with Uncertainty **certainty factors Explanation Ease of Modification Transportability Adaptive learning @@Advantages: Capture of scarce expertise Superior problem solving Reliability Work with incomplete information Transfer of knowledge
Marcus was a man. man(Marcus) 2. Marcus was a Pompeian.
Pompeian(Marcus)3. All Pompeians were Romans. ∀x: Pompeian(x) → Roman(x) 4. Caesar was a ruler. ruler(Caesar) 5. All Romans were either loyal to Caesar or hated him. ∀x: Roman(x) → loyalto(X. Caesar) V hate(x, Caesar) 6. Everyone is loyal to someone. ∀x : ∃ y : Ioyalto(x,y) 7. People only try to assassinate rulers they are not loyal to. ∀ x : ∀ y : person(x) ∧ ruler(y) ∧ tryassassinate(x,y) → ¬ Ioyalto(x,y) 8. Marcus tried to assassinate Caesar. tryassassinate (Marcus, Caesar) 9. All men are people. ∀x : man(x) → person(x) AI:Thinking humanly•Acting humanly•Thinking rationally•Acting rationally Cognitive science: the brain as an information processing machineRequires scientific theories of how the brain works How to understand cognition as a computational process?Introspection: try to think about how we think Predict and test behavior of human subjects Image the brain, examine neurological data •The latter two methodologies are the domains of cognitive science and cognitive neuroscience Turing (1950) "Computing machinery and intelligence" •The Turing Test What capabilities would a computer need to have to pass the Turing Test?Natural language processing Knowledge representation *Automated reasoning*Machine learning Turing predicted that by the year 2000, machines would be able to fool 30% of human judges for five minutes SearchUninformed search, informed search Adversarial search: minimax Constraint satisfaction problems *Planning•Logic•ProbabilityBasic laws of probability*Bayes networks*Hidden Markov Models•LearningDecision trees Linear classifiers: neural nets, support vector machines Reinforcement learning @@History of ai: Early excitement 1940s McCulloch & Pitts neurons; Hebb’s learning rule 1950 Turing’s “Computing Machinery and Intelligence” 1954 Georgetown-IBM machine translation experiment 1956 Dartmouth meeting: “Artificial Intelligence” adopted 1950s-1960s “Look, Ma, no hands!” period: Samuel’s checkers program, Newell & Simon’s Logic Theorist, Gelernter’s Geometry Engine The rest of the story 1974-1980 The first “AI winter” 1970s Knowledge-based approaches 1980-88 Expert systems boom 1988-93 Expert system bust; the second “AI winter” 1986 Neural networks return to popularity 1988 Pearl’s Probabilistic Reasoning in Intelligent Systems 1990 Backlash against symbolic systems; Brooks’ “nouvelle AI” 1995-present Increasing specialization of the field Agent-based systems Machine learning everywhere Tackling general intelligence again? @@Genetic algorithms: (GAs) are a technique to solve problems which need optimization.*GAs are a subclass of Evolutionary Computing and are random search algorithms.*GAs are based on Darwin’s theory of evolution.*History of GAs: @@Searcg space:*Evolutionary computing evolved in the 1960s.*GAs were created by John Holland in the mid-1970s. Most often one is looking for the best solution in a specific subset of solutions.*This subset is called the search space (or state space).*Every point in the search space is a possible solution.*Therefore every point has a fitness value, depending on the problem definition.*GAs are used to search the search space for the best solution, e.g. a minimum.*Difficulties are the local minima and the starting point of the search.@@@basic algo: Starting with a subset of n randomly chosen solutions from the search space (i.e. chromosomes). This is the population.*This population is used to produce a next generation of individuals by reproduction.*Individuals with a higher fitness have more chance to reproduce (i.e. natural selection). Outline of the basic algorithm 0 START : Create random population of n chromosomes1 FITNESS : Evaluate fitness f(x) of each chromosome in the population 2 NEW POPULATION 1 REPRODUCTION/SELECTION : Based on f(x) 2 CROSS OVER : Cross-over chromosomes 3 MUTATION : Mutate chromosome 3 REPLACE : Replace old with new population: the new generation 4 TEST : Test problem criterium 5 LOOP : Continue step 1 – 4 untill criterium is satisfied Choose n random crossover points. Split along those points. Glue parts, alternating between parents. Generalization of 1 point. Ant Colony Optimization Algorithms:*Construction heuristics •How ants find shortest route •Stigmergy •General ACO metaheuristic •Ant System for TSP @@Motivation •NP-hard problems – no algorithms that could solve large instances of these problems to optimality Discrete combinatory problems•Approximate metods – can find solutions of good quality in reasonable time *Approximate metods Local search/optimization •Iteratively improves a complete solution (typically initialized at random) till it reaches some local optimum. Construction algorithms •Build a solution making use of some problem-specific heuristic information •Ant Colony Optimization (ACO) algorithms – extend traditional construction heuristics with an ability to exploit experience gathered during the optimization process@@Construction Algorithms •Build solutions to a problem under consideration in an incremental way starting with an empty initial solution and iteratively adding opportunely defined solution components without backtracking until a complete solution is obtained. Procedure GreedyConstructionHeur end return sp end •Pros/Cons + fast, solutions of reasonable quality - Solution may be far from optimum -Generate only limited number of different solutions -Decisions made at early stages reduce a set of possible steps at latter stages @@Ant Algorithms: Biological Inspiration •Inspired by behavior of an ant colony Social insects – behave towards survival of the colony Simple individual behavior complex behavior of a colony•Ability to find the shortest path from the colony to the source of food and back using an indirect communication via pheromone •Write – ants lay down pheromone on their way to food •Read – ant detects pheromone (can sense different intensity) laid down by other ants and can choose a direction of the highest concentration of pheromone. •Emergence – this simple behavior applied by the whole colony can lead to emergence of the shortest path @@Ant Colony Optimization Metaheuristic •ACO can be applied to any discrete optimization problem for which some solution construction mechanism can be conceived. •Artificial ants are stochastic solution construction heuristics that probabilistically build a solution by iteratively adding solution components to partial solutions by taking into account heuristic information on the problem instance being solved, if available, (artificial) pheromone trails which change dynamically at run-time to reflect the agents’ acquired search experience. •Stochastic component allows generating a large number of different solutions. Knowledge Representation: Knowledge facts/principles (info. must be True) intelligence procedural knowledge algebraic eq.Semantic Network composed of multiple attribute values *Design Hierarchical structure Ex: (tree structure for below)*All Bikes are 2- wheelers*All 2-wheelers are Vehicles*All 4-wheelers are Vehicles*All cars are 4-wheelers*Kavasaki is a gear bike*All gear bikes are 2-wheelers *Knowledge acquisition acquire a KB thro domain expert*Select domain – knowledge engineers – experts*Inference engine control mechanism for expert system Converting to clausal form: 1. man (Marcus). 2. Pompeian (Marcus). 3. ¬(Pompeian(x1) ) V Roman(x1)). Removing implication and universal quantifier by replacing it with a variable x1 4. ruler (Caesar). 5. ¬ (Roman(x2) ) V loyalto(x2,Caesar ) V hate(x2,Caesar )). Removing implication and universal quantifier by replacing it with a variable x2 6. loyalto(x3, F(x3)). Removing universal quantifier by replacing it with a variable x3 and existential quantifier by F(x3) since existential quantifier was inside universal, that’s why a function 7. ¬ man(x4) V ¬ruler (y1) V ¬tryassassinate(x4,y1) V ¬loyalto(x,y)). Removing universal quantifier for x and y by replacing it with x4 and y1 respectively. 8. tryassassinate(Marcus, Caesar ) 9. Question: ¬hate(Marcus; Ceasar ) Proof by Refutation Event A1. It rains on Marie's wedding.*Event A2. It does not rain on Marie's wedding.*Event B. The weatherman predicts rain. In terms of probabilities, we know the following: P( A1 ) = 5/365 =0.0136985 [It rains 5 days out of the year.]P( A2 ) = 360/365 = 0.9863014 [It does not rain 360 days out of the year.]P( B | A1 ) = 0.9 [When it rains, the weatherman predicts rain 90% of the time.]P( B | A2 ) = 0.1 [When it does not rain, the weatherman predicts rain 10% of the time.] We want to know P( A1 | B ), the probability it will rain on the day of Marie's wedding, given a forecast for rain by the weatherman. The answer can be determined from Bayes' theorem, as shown below. P( A1 ) P( B | A1 ) P( A1 | B ) = P( A1 ) P( B | A1 ) + P( A2 ) P( B | A2 ) P( A1 | B ) = (0.014)(0.9) / [ (0.014)(0.9) + (0.986)(0.1) ] P( A1 | B ) = 0.111