Techniques and Applications of AI

Intelligence requires knowledge. Intelligence is the ability to store , retrieve and act on data – efficiently and effectively. Knowledge is a collection of ‘facts’. To manipulate these facts by a program, a suitable representation is required. A good representation facilitates problem solving. Artificial Intelligence enable computers to recognize the knowledge , reason ( questioning it to find out the logic) and act upon it. It is quite difficult to define the characteristics of Knowledge correctly. It differs from data by being organized in a way that corresponds to its application.

Learning is not very different from perception, they both find out about the world based on experience. Perception is the short time scale (where am I?) .Learning is the long time scale (what’s the coefficient of friction?).An AI technique is a method that exploits knowledge and captures generalizations that share properties, are grouped together, rather than being allowed separate representation.AI Techniques tell us how we represent, manipulate and reason with knowledge in order to solve problems. Some of these techniques employ an agent whose job is to use sequences of percepts to estimate the missing details in the world of dynamics.

Different techniques in AI are as follows :

Branch and Bound :

Here, the domain is classified into sections. Knowledge is stored in decision trees . Tree of decisions is formed ( conditions with “yes” or “no” branches). It is a semi-transparent technique. With large problems , it becomes difficult for human interpretation.

Application : Insurance fraud detection & Credit assessment of the loan application for processing bank loans.

Guided search :

It is mainly used in the Optimisation (maximization) problems . Here, the domain is described by a function and optimisation techniques  do not actually produce only local optima.This does not search previously visited point. Therefore , it will not become stuck or get into a loop forever.

Application :Finding the optimal solution with maximum weight , Machine learning applications.

Heuristic search :  

A heuristic is a way of trying to discover an idea. Heuristics are rules of thumb and they do not guarantee a solution to a problem. Heuristic functions are used in some approaches to search or to measure how far a node in a search tree seems to be from a goal. Heuristic technique compares two nodes in a search tree to see if one is better than the other, i.e. constitutes an advance toward the goal, and may be more useful. These algorithms appear to be really intelligent because they perform better. Heuristic algorithms are more efficient because they take advantage of feedback from the data to direct the search path. This search technique is a weak one but can be effective if applied correctly and  it requires domain specific information. Heuristics are knowledge about domain, which help search and reasoning in its domain. Heuristic is a function that, when applied to a state, returns value as estimated merit of state, with respect to goal.  Heuristic evaluation function estimates likelihood of given state leading to goal state.  Heuristic search function estimates cost from current state to goal, presuming function is efficient.

Inference Engine : This technique uses efficient procedures and rules . It is essential in deducting a correct, flawless solution. In case of knowledge-based ES, the Inference Engine acquires and manipulates the knowledge from the knowledge base to arrive at a particular solution. In case of rule based ES, it −  Applies rules repeatedly to the facts, which are obtained from earlier rule application.  Adds new knowledge into the knowledge base if required.  Resolves rules conflict when multiple rules are applicable to a particular case. To recommend a solution, the Inference Engine uses the following strategies −  Forward & Backward Chaining.

Forward Chaining – It is a strategy of an expert system to answer the question, “What can happen next?” Here, the Inference Engine follows the chain of conditions and derivations and finally deduces the outcome. It considers all the facts and rules, and sorts them before concluding to a solution. This strategy is followed for working on conclusion, result, or effect. For example, prediction of share market status as an effect of changes in interest rates.

Backward Chaining -With this strategy, an expert system finds out the answer to the question, “Why this happened?” On the basis of what has already happened, the Inference Engine tries to find out which conditions could have happened in the past for this result. This strategy is followed for finding out cause or reason. For example, diagnosis of blood cancer in humans.

Genetic programming :  Genetic programming is an automated method for creating a working computer program from a high-level problem statement of a problem. Genetic programming starts from a high-level statement of ‘what needs to be done’ and automatically creates a computer program to solve the problem. Supervised learning is the machine learning task of inferring a function from labeled training data. In supervised learning, each example is a pair consisting of an input object (typically a vector) and the desired output value (also called the supervisory signal ).

Neural Networks – Backward Propagation :

Neural Networks model a brain. Nodes represent neurons .They are linked to other nodes via connections ( representing synapses). Nodes send messages to their output when a threshold from their inputs has been reached. Neural networks typically take a vector of input values and produce a vector of output values. Inside, they train weights of “neurons”. Neural networks use supervised learning, in which inputs and outputs are known and the goal is to build a representation of a function that will approximate the input to output mapping.

This approach is learning by example―given a set of right answers, it learns the general patterns. Reinforcement Learning  learns by experience―given some set of actions and an eventual reward or punishment, it learns which actions are good or bad.It is a “Black Box” technique.

Applications : Modelling of industrial systems , Speech recognition programs.

Expert systems :

It is a transparent technique.It becomes difficult for human interpretation with very large problems – above 1000 rules, the logic chain becomes huge.

Applications : Process monitoring and control in medical fields , chemical industries etc.

Summing up….

An AI technique is a method that exploits knowledge and captures generalizations that share properties and can be grouped together. AI Techniques tell us how we represent, manipulate and reason with knowledge in order to solve problems.These techniques have practical applications in almost every field. It forms the basic foundation of the new revolutionizing technology – Intelligent connectivity.

Kindly check the following link to explore further about the topics :