Some definitions/pioneering thoughts on Artificial Intelligence…


In Computer Science, work termed “AI” has traditionally focused on the high-level problem; on imparting high-level abilities to “use language, form abstractions and concepts” and to “solve kinds of problems now reserved for humans” (McCarthy et al. 1955)

AI is the science of making computers do things that require intelligence like humans (Minsky)

The automation of activities that we associate with human thinking, activities such as decision making, problem solving, learning … (Bellman, 1978)

Physicists ask what kind of place this Universe is and seek to characterize its behavior systematically. Biologists ask what it means for a physical system to be living. We (in AI) wonder what kind of information processing system can ask such questions – Avron Barr and Edward Feigenbaum (1981)

The fundamental goal of this research is not merely to mimic intelligence or produce some clever fake. “AI” wants the genuine article; machines with minds – John Haugeland (1985)

AI is the study of mental faculties through the use of computational models – Eugene Charnaik and Drew McDermott (1985)

We call programs ‘intelligent’ if they exhibit behaviors that would be regarded intelligent if they were exhibited by human beings – Herbert Simon

AI is the study of techniques for solving exponentially hard problems in polynomial time by exploiting knowledge about the problem domain – Elaine Rich

The art of creating machines that perform functions that require intelligence when performed by people (Kurzweil, 1990)

The study of the computation that make it possible to perceive, reason and act (Winston, 1992)

AI … is concerned with intelligent behavior in artifacts (Nilsson, 1998)

Computational Intelligence is the study of the design of intelligent agents (Poole et al., 1998)

AI is a branch of computer science concerned with the study and creation of computer systems that exhibit some form of intelligence: systems that learn new concepts and tasks, systems that can reason and draw useful conclusions about the world around us, systems that can understand a natural language or perceive and comprehend a visual scene, and systems that perform other types of feats that require human types of intelligence (Dan W. Patterson)



  1. Deepak Khemani, A first course in AI, Mcgraw Hill Education
  2. Stuart Russell and Peter Norvig, AI: a modern approach, Second Edition, Pearson
  4. Dan W. Patterson, Introduction to Artificial Intelligence and Expert Systems, PHI ISBN-978-81-203-0777-3

Leave a comment

Filed under artificial intelligence

About KBS, Knowledge Engineering and Heuristics …

  1. What are Knowledge Based Systems?

Knowledge Based systems are systems that contain a good amount of knowledge to perform difficult tasks. In his seminal 1977 paper at the Joint International Conference on Artificial Intelligence (IJCAI), Edward Feigenbaum emphasized that the real power of an expert system comes from the knowledge it possesses rather than the particular inference schemes and other formalisms it employs [1].

Much of the work in AI has been related to Knowledge Based systems, which includes work in Expert Systems, Natural Language Understanding, Vision and others.

Knowledge based systems derive their power from the knowledge base. The knowledge base is a repository of facts, rules, heuristics and procedures. The knowledge base is segregated from the control and inferencing components enabling it to add additional knowledge or refine existing knowledge independently.


  1. Dan W. Patterson, Introduction to AI and Expert Systems, PHI, ISBN-978-81-203-0777-3
  2. What is Knowledge Engineering?

Knowledge Engineering is a skill-set expected in a Knowledge Engineer. A knowledge engineer is responsible for designing and building a KBS (Knowledge based systems). For this, a knowledge engineer has to interact with human experts (domain knowledge) and incorporate the knowledge using Knowledge Representation schemes (rules, frames & others) into a knowledge base. Once appropriate knowledge have been encoded, a KBS system can be useful for discovery of knowledge to be made use of in real-life applications such as Robotics, Tele Medicine or in Traffic Control.

The success of Knowledge Engineering would depend on the ability of the Knowledge Engineer to extract expert knowledge from the minds of human experts and other informative sources from relevant domains.

The AI scientist Edward Feigenbaum describes the knowledge engineer in the following words “The Knowledge Engineer practices the art of bringing the principles and tools of AI research to bear on difficult application problems requiring experts’ knowledge for their solution. The technical issues of acquiring this knowledge, representing it, and using it appropriately to construct and explain lines-of-reasoning, are important problems in the design of knowledge-based systems … The art of constructing intelligent agents is both a part of and an extension of the programming art. It is the art of building complex computer programs that represent and reason with knowledge of the world (Feigenbaum, 1977, p. 1015)” [1]


  1. George M. Marakas, Decision Support Systems in the 21st Century, PHI ISBN-978-81-203-2376-6


  1. What is a Heuristic? Justify how heuristics can be effective to solve problems?

A heuristic is an intelligent guess used in problem solving. It is an approximation done to reduce the search space. A heuristic function defines a state in terms of a number and this number is used for decision making in the search problems. Heuristics are employed when

  • The perfect solution to the problem is not known
  • The best solution is not computationally feasible

Example – In a best first search algorithm the heuristic used is the distance of the node from the goal state. In the 8 tile problem, hamming distance and Manhattan distance are used as heuristics

Effectiveness of heuristics – Heuristics are effective to solve search problems because they drastically reduce the search space because each time the alternatives are explored one out of them is chosen which is further expanded, so we do not have to traverse all nodes.

Leave a comment

Filed under artificial intelligence



Q1. List  the major AI technologies

  • Robotics
  • Multimedia Recognition systems
  • Expert Systems
  • Gaming
  • Virtual Playmate
  • Virtual Assistant (Siri, Google Now)
  • Automated Vehicles
  • Natural Language Processing

Q2. What do you understand by learning?       

Learning is a process of acquiring knowledge and storing it for future use. Storing knowledge can help in making use of it when a similar situation in problem solving arises later.

Knowledge based systems are dependent on learning. They interact with the environment to learn and acquire knowledge fed into them by human experts. Learning involves using existing knowledge in problem solving and gaining newer insights while doing so.

Q3. Explain briefly: Reasoning under Uncertainty

The capability of using existing knowledge to make an inference when more than one possible outcome is possible e.g. ,thinking of taking a left or a right turn while driving when suddenly a dog appears on the road

Q4. What are the skills needed to make a good Knowledge Based System?

Following are some of the skills

  • Good knowledge acquisition methods/devices
  • Ability to represent knowledge such that utility is served. Raw data and facts need to be structured and organized effectively so that inferences and retrieval can take place in an easy manner
  • Building functions to create line-of-reasoning
  • Collate knowledge from multiple human experts to incorporate domain knowledge in knowledge base
  • Adaptive learning – As newer knowledge emerges, older body may require a re-look and reorientation to fit with the latest knowledge
  • Good inference engines and rules, good reasoning…

Q5. Explain backward chaining with an example

In backward chaining we go backwards from goal to generate facts or rules which can lead us to achieve that particular goal. For example, in the case of GPS, we enter our destination (goal) and the GPS system finds out the best route (facts or rules) so that the given destination can be reached.

Leave a comment

Filed under artificial intelligence

What is Knowledge?

Knowledge is an application of a combination of instincts, ideas, rules, procedures and information to guide the actions and decisions of a problem solver within a particular problem context. In this sense, knowledge is an interpretation made by the mind [2].

Knowledge is an asset that exists in intelligent beings accrued over the years that help perform or solve a complex task. It may constitute a set of patterns and correlations/associations among data/events that facilitate deft problem solving. Knowledge may be acquired over years through education, peer and self learning, attending conferences and workshops and Intelligent Analysis of prevailing facts and literature.

Knowledge based problem solving would require intelligent heuristic based methods besides adequate command over domain area. Capability to make good decisions under uncertain situations would be needed. Adaptive learning ability is essential with the dynamic nature of knowledge seen in recent times.

Let us look at this hypothetical situation. Say the Indian Govt. decides to elevate the standard of Indian football to International class. As a preliminary exercise, it selects the best thirty players of the country. Pele, the great footballer, is recruited as an adviser/coach to assess the capability of the selected players. To determine whether Indian footballers can compete at the highest level of International football, Pele would need his vast knowledge – his experiences from myriad spheres of having played brilliant International soccer, the techniques and skills needed to make the cut, the need of physical attributes like physique and stamina for the players. He can arrive at a conclusion only after that.

Finally, an overall picture of knowledge cannot be complete without also knowing the meaning of closely related concepts such as understanding, learning, thinking, remembering and reasoning [1].


1.     Dan W. Patterson, Introduction to AI and Expert Systems, PHI, ISBN-978-81-203-0777-

2.    George M. Marakas, Decision Support Systems in the 21st Century, PHI ISBN-978-81-203-2376-6

Leave a comment

Filed under artificial intelligence

Applications of Differential Equations in Computer Science

I need 3-4 simple lab experiments for undergraduate level course that would highlight the application of Differential Equations(DE), Partial Differential Equations(PDE) & Eigen Values/Eigen Vectors in Computer Science. DE is used in gradient descent in Back Propagation Neural Network and in SVM (Support Vector Machines)but this is likely to prove difficult for students undergoing a Maths course unfamiliar with AI/NN. Any suggestions/help would be greatly appreciated …

Leave a comment

Filed under Articles, General

Does a computer really think? [2]

For decade now the proponents of ‘Strong AI’  have tried to persuade us that it is only a matter of century or two (some have lowered the time to fifty years!) until electronic computers will be doing everything a human mind can do. Stimulated by science fiction read in their youth, and convinced that our minds are simply ‘computers made of meat’ (as Marvin Minsky once put it), they take for granted that pleasure and pain, the appreciation of beauty and humour, consciousness, and free will are capacities that will emerge naturally when electronic robots become sufficiently complex in their algorithmic behavior [1].

Some philosophers of science notably John Searle strongly disagree. To them a computer is not essentially different from mechanical calculators that operate with wheels, levers, or anything that transmit signals. Because electricity travels through wires faster than other forms of energy (except light) it can twiddle symbols more rapidly than mechanical calculators, and therefore handle tasks of enormous complexity. But does an electrical computer understand what is it doing? [1].

There is no dispute over computers exhibiting intelligent behavior in many situation. The bedrock of manifestation of intelligence by computers is its knowledge bases and inference mechanism, heuristic and optimizer based search approaches in lessening computational complexity. However most AI programs use pattern recognition and search techniques which lead one to believe that these are in essence non-intelligent in nature. One need to provide a lot of information to the computer to initiate a modicum of learning in it. Only thereafter the computer may manifest some degree of intelligence in task solving.

However the human mind is just too complex to duplicate. Computers certainly cannot think in the same way humans do but they can be very useful for increasing our productivity. This is done by several commercial AI technologies [2].



1.     Roger Penrose, The Emperor’s New Mind, Penguin Books, Chapter 1

2.    Turban, Rainer, Potter, Introduction to Information Technology second edition, Pg 394, Wiley India ISBN 978-81-265-0968-3

Leave a comment

Filed under Articles, artificial intelligence

What are the challenges for building intelligent systems?


It is an acknowledged fact that to build an effective intelligent system one would require a lot of data. Learning won’t be effective with a small data size. Greater the volume of data, the better is the learning. For example, the successful intelligent based application emerged in those domains where data is quite readily available e.g. Image Recognition system. Computer requires several orders of data in excess than humans to perform the same learning task. The owners of data are the big companies like Google, Amazon, Facebook, Microsoft and a few more. These organizations do have an edge over others to build intelligent applications.

The basis of any intelligent decision making is knowledge. How to capture knowledge, represent and codify it for later use in inference making/reasoning is a non trivial task. We are still far, far away from building Automated Real Time Reasoning system. Current Machine Learning techniques like Decision Tree and Neural Networks have obtained limited success in a narrow domain. Even Neural learning is based on synaptic weight changes based on Hebb’s work which dates back to 1959 …

Leave a comment

Filed under Articles, artificial intelligence