Week 5

• what is the study of economic decision-making and the factors that affect it?

  behavioural economics
• behavioural economics applies scientific methods to understand the cognitive, emotional and social influences on the way we think about things such as money, goods, services, buying and selling
• the classical view of behavioural economics is that people are rational and always make choices that increase their money or value
• the ultimatum game: a game where participants are given a choice between two options, one of which is better than the other (splitting $1000 with someone)
• loss aversion: the principle by which people are more sensitive to losses than to equivalent gains (the flu outbreak example)
• loss aversion and framing effect shows that we prefer choices that are framed positively rather than negatively ($5 discount vs. avoid $5 surcharge)
• mental accounting: principle by which we reason differently about money depending on the overall amounts (phone priced down a few blocks away example)
• endowment effect: tendency to stay with what we have and to prefer it over other options (investments example)
• sunk-cost fallacy: the result of our tendency to stick with those things we’ve already put a lot of time, energy, or money into although letting those go would be financially more reasonable (chris and his old car example)
• methodological variety: attempt to develop objective science of behaviour
• the goal of work in artificial intelligence is to build machines that perform tasks normally requiring human intelligence
• AI and cognitive science: explore how humans do such intelligent tasks and develop algorithms for given machine capabilities
• true or false: the more one knows, the more intelligent that person becomes?

  false
• the ability to learn from experience, acquire new knowledge and apply the best techniques to meet our goals is what makes us intelligence
• which AI technique finds all the links to the answer (many different algorithms exist and the right one is chosen based on the problem, data, constraints and available resources)?

  search
• which AI technique gets feedback from previous work and tries to reduce error (machine learning methods)?

  learn
• which AI technique generates usable knowledge from data by analytics (statistical and machine learning methods)?

  knowledge acquisition/retrieval
• which AI technique quickly finds the best decision where methods are rule-based, case-based, hybrid systems or apply multiple constraints satisfaction techniques?

  decision making
• cognitive simulation: simulate processes of the mind
• what is step 1 of cognitive simulation?

  find how humans do it
• what is step 2 of cognitive simulation?

  form a hypothesis
• what is step 3 of cognitive simulation?

  validate using human participants
• what is step 4 of cognitive simulation?

  implement it using the machine
• intelligent agents: a computer program that can learn from experience and make decisions based on its knowledge
• what is this?

  turing's finite state machine (FSM)

  
• what is this?

  cognitive/IA model - kenneth craik

  
• rule-based: decisions are guided by rules (e.g., expert systems)
• case-based: decisions are guided by example cases or heuristics
• expert systems employ knowledge (facts), logic (rules, in the form of an if-then structure), and algorithms or procedure for searching and choosing rules
• blind searches contain no information about the layout of the problem space (an exhaustive search inspects each state to find a goal state)
• a depth-first solution follows a search path to the end before moving on to another path
• breadth-first search checks all the states you can reach in one step from the current position and then select one in a sequence and explore all the states you can get to next
• evolutionary computing: applies concepts such as the fittest solution survives and reproduction algorithms generate new solutions
• artificial neural network: modelled after the brain neural system where information comes from many sources and their interconnections and importance decide the final outcome
• understanding a natural language involves an individual’s assimilation of linguistic expression in some form (speech or writing) extracting its meaning and undertaking some action that constitutes a response to this meaning
• what stage of natural language processing is where the acoustic speech signal is analyzed to determine the sequence of spoken words?

  speech recognition
• what stage of natural language processing is where word sequence is analyzed via the use of knowledge of the language’s grammar?

  syntactic analysis
• what stage of natural language processing is where the sentence structure and meanings of the words are used to derive a partial representation of the meaning of a sentence?

  semantic analysis
• what stage of natural language processing is where it produces a complete meaning for the sentence via the application of contextual information?

  pragmatic analysis
• a chatterbot (talkbot, chatbot) is a computer program which conducts a conversation via auditory or textual methods