AI-lowres

  Artificial intelligence is the most exciting field nowadays and I came to recognise it while research more about this area.Courtesy to university of California,their spring course inspired me to write this.Moreover we watched so many films related to AI like The Matrix,The terminator etc. So a normal human being is not at all a newbie to this area. We all have intelligence,ie,we have the capacity to learn and solve problems,act rationally …we have the ability to interact with real world,reasoning and planning, learning and adaptation.We can relate almost all disciplines with AI. More details about it available in wikipedia. ‎ 

Our brain which is complicated than we thought,so yes we can expect computers with as many basic processing elements as our brain in the near future.We can google a lot of success stories like chess playing ,is a classic AI problem,and is definitely well defined one.We need to be aware of many basic things before get into this area.Some are discussed here: Can computers talk?Speech synthesis,which translates text to phonetic form and use pronunciation rules to map phonemes to actual sound.But it has many difficulties like lack of naturalism(some systems handle this properly),emphasis,emotion etc are much harder,ie,humans understand what they are saying - machines don’t. Like speech synthesis, recognition of human speech is much more difficult.Since speech is continuous,have large vocabularies, background noise etc After recognition do you think Computers can understand speech:No, much of what we say is beyond the capabilities of a computer to understand at present. While coming to the point learning and adaptation,,it is YES !! Consider a computer learning to drive in a freeway, we can teach it what to do(rules and all).The area Machine learning (search for more details) allows computer to learn to do things without explicit programming,these are all conclude that computers can learn and adapt,when presented with information in the appropriate way. Now lets deal with “see”, can computers see?Humans can effortlessly recognize objects,then why visual recognition a hard problem to machines.Mostly the conclusion to this part is NO,computers can only “see” certain types of objects under limited circumstances. But in case of certain constrained problems like face recognition, it is YES. This is little more hard for machines that to plan and make optimal decisions.For that it needs intelligence,which involves solving problems and making decisions and plans.What makes planning hard? It is of course of the reason that the world is not predictable so AI system only successful in constrained planning problems.We can conclude that real-world planning and decision making is still beyond the capabilities of modern computers. AI can be modelled in different ways like exactly how humans actually think,act ,…Do you think that it is important that an intelligent system act like a human?For thinking like humanly we need a cognitive approach:try to get “inside “ our minds.Problem is that we don’t behave rationally,so reverse engineering is hard to do.And the funny thing is that brain’s hardware is very different to a computer program. The most seen implementation in these years is Robots and AI. Both terms are hard to define.Some modern robots have the ability to learn in a limited capacity.You can get a lot of articles,inspiring stories/ experiences related to this area in web.It is really interesting to connect how and why we learn and think, and experimenting these ideas using robots.

** Many research problems remaining in this AI area and we can expect more intelligent systems in the near future.!**

  • link: https://rosnikv92.wordpress.com/2013/07/03/will-artificial-intelligence-rule-the-world-or-not/ author: rosnikv description: post_id: 139 created: 2013/07/03 17:32:56 created_gmt: 2013/07/03 17:32:56 comment_status: open post_name: will-artificial-intelligence-rule-the-world-or-not status: publish post_type: post *