Over the past few years, you might have come across the term artificial intelligence and have imagined it to be a vivid personification of extraterrestrial beings or robots. But, what is artificial intelligence in fact? In this section, we will explore the meaning and semantics of the term artificial intelligence.

Artificial intelligence is the search for a way to map intelligence into mechanical hardware and enable a structure into that system to formalize thought. No formal definition, as yet, is available for as to what artificial intelligence actually is. Over the course of this section, we will try to formulate a working definition, reasoning and articulating facts and preferences of various other authors and practitioners of the field. To start with, we would think of the word artificial and intelligence as main sources of inspiration and come up with a brief description of the same as follows:

With our first set of formal declaration of the concept, we tread gradually through different semantics of it and try to explore a much broader definition for AI. In their book Artificial Intelligence: A Modern Approach, authors Russell and Norvig tried to establish a clear classification of the definition of the field into distinct categories based on working definitions from other authors commenting on AI. The demarcation of concepts holds true to these clauses for systems that:

So, one would be tempted to improve upon the definition given above to include these facts into perspective such that the definition that we end up with says that:

There are numerous definitions of what artificial intelligence is.

We end up with four possible goals:

What is rationality? -> "doing the right thing" - simply speaking.

Definitions:

For (1.): "The art of creating machines that perform functions that require intelligence when performed by humans" (Kurzweil). Involves cognitive modeling - we have to determine how humans think in a literal sense (explain the inner workings of the human mind, which requires experimental inspection or psychological testing)

For (2.): "GPS - General Problem Solver" (Newell and Simon). Deals with "right thinking" and dives into the field of logic. Uses logic to represent the world and relationships between objects in it and come to conclusions about it. Problems: hard to encode informal knowledge into a formal logic system and theorem provers have limitations (if there's no solution to a given logical notation).

For (3.): Turing defined intelligent behavior as the ability to achieve human-level performance in all cognitive tasks, sufficient to fool a human person (Turing Test).Physical contact to the machine has to be avoided, because physical appearance is not relevant to exhibit intelligence. However, the "Total Turing Test" includes appearance by encompassing visual input and robotics as well.

For (4.): The rational agent - achieving one's goals given one's beliefs. Instead of focusing on humans, this approach is more general, focusing on agents (which perceive and act). More general than strict logical approach (i.e. thinking rationally).

A human judge engages in a natural language conversation with two other parties, one a human and the other a machine; if the judge cannot reliably tell which is which, then the machine is said to pass the test. It is assumed that both the human and the machine try to appear human.

Tried to debunk the claims of "Strong AI". Searle opposed the view that the human mind IS a computer and that consequently, any appropriately construed computer program could also be a mind.

A man who doesn't speak Chinese sits in a room and is handed sheets of paper with Chinese symbols through a slit. He has access to a manual with a comprehensive set of rules on how to respond to this input. Searle argues, that even though he can answer correctly, he doesn't understand Chinese; he's just following syntactical rules and doesn't have access to the meaning of them. Searle believes that "mind" emerges from brain functions, but is more than a "computer program" - it requires intentionality, consciousness, subjectivity and mental causation.

The problem deals with the relationship between mental and physical events. Suppose I decide to stand up. My decision could be seen as a mental event. Now something happens in my brain (neurons firing, chemicals flowing, you name it) which could be seen as a physical, measurable event. How can it be, that something "mental" causes something "physical". Hence it's hard to claim that both are completely different things. One could go on and claim that these mental events are, in fact, physical events: My decision is neurons firing, but I'm not aware of this - I feel like I made a decision independently from the physical. Of course this works the other way round too: everything physical could, in fact, be mental events. When I stand up after having made the decision (mental event), I'm not physically standing up, but my actions cause mental events in my and the bystanders minds - physical reality is an illusion.

It is the question of how to determine which things remain the same in a changing environment.

Occurs whenever an organism or artificial intelligence is at some current state and does not know how to proceed in order to reach a desired goal state. This is considered to be a problem that can be solved by coming up with a series of actions that lead to the goal state (the "solving").

In general, search is an algorithm that takes a problem as input and returns with a solution from the searchspace. The search space is the set of all possible solutions. We dealt a lot with so called "state space search" where the problem is to find a goal state or a path from some initial state to a goal state in the state space. A state space is a collection of states, arcs between them and a non-empty set of start states and goal states. It is helpful to think of the search as building up a search tree - from any given node (state): what are my options to go next (towards the goal), eventually reaching the goal.

Uninformed search (blind search) has no information about the number of steps or the path costs from the current state to the goal. They can only distinguish a goal state from a non-goal state. There is no bias to go towards the desired goal.

For search algorithms, open list usually means the set of nodes yet to be explored and closed list the set of nodes that have already been explored.

In informed search, a heuristic is used as a guide that leads to better overall performance in getting to the goal state. Instead of exploring the search tree blindly, one node at a time, the nodes that we could go to are ordered according to some evaluation function that determines which node is probably the "best" to go to next. This node is then expanded and the process is repeated (i.e. Best First Search). A* Search is a form of BestFS. In order to direct the search towards the goal, the evaluation function must include some estimate of the cost to reach the closest goal state from a given state. This can be based on knowledge about the problem domain, the description of the current node, the search cost up to the current node BestFS optimizes DFS by expanding the most promising node first. Efficient selection of the current best candidate is realized by a priority queue.

Finding Heuristics:

Minimax is for deterministic games with perfect information. Non-Deterministic games will use the expectiminimax algorithm.

For CSPs, states in the search space are defined by the values of a set of variables, which can get assigned a value from a specific domain, and the goal test is a set of constraints that the variables must obey in order to make up a valid solution to the initial problem.

Example: 8-queens problem; variables could be the positions of each of the eight queens, the constraint to pass the goal test is that no two queens can be such that they harm each other.

Different types of constraints:

In addition to that, constraints can be absolute or preference (the former rules out certain solutions, the latter just says which solutions are preferred). The domain can be discrete or continuous. In each step of the search, it is checked if any variable has violated one of the constraints - if yes: backtrack and rule out this path.

Forward checking checks if any decisions on yet unassigned variables would be ruled out by assigning a value to a variable. It deletes any conflicting values from the set of possible values for each of the unassigned variables. If one of the sets becomes empty - backtrack immediately.

There are also heuristics for making decisions on variable assignments.

Selecting the next variable:

After selecting the variable:

variables connected to the current variable through constraints.

CSP's that work with iterative improvement are often called "heuristic repair" methods, because they repair inconsistencies in the current configuration. Tree-structured CSPs can be solved in linear time.

Visit link:

Artificial Intelligence/Definition - Wikibooks, open books ...