Artificial Intelligence

   

Cognitive Architecture for Personable and Human-Like ai :A Perspective

Authors: Arvind Chitra Rajasekaran

In this article we will introduce a cognitive architecture for creating a more human like and personable artificial intelligence. Recent works such as those by Marvin Minsky, Google DeepMind and cognitive models like AMBR, DUAL that aim to propose/discover an approach to commonsense AI have been promising, since they show that human intelligence can be emulated with a divide and conquer approach on a machine. These frameworks work with an universal model of the human mind and do not account for the variability between human beings. It is these differences between human beings that make communication possible and gives them a sense of identity. Thus, this work, despite being grounded in these methods, will differ in hypothesizing machines that are diverse in their behavior compared to each other and have the ability to express a dynamic personality like a human being. To achieve such individuality in machines, we characterize the various aspects that can be dynamically programmed onto a machine by its human owners. In order to ensure this on a scale parallel to how humans develop their individuality, we first assume a child-like intelligence in a machine that is more malleable and which then develops into a more concrete, mature version. By having a set of tunable inner parameters called aspects which respond to external stimuli from their human owners, machines can achieve personability. The result of this work would be that we will not only be able to bond with the intelligent machines and relate to them in a friendly way, we will also be able to perceive them as having a personality, and that they have their limitations. Just as each human being is unique, we will have machines that are unique and individualistic. We will see how they can achieve intuition, and a drive to find meaning in life, all of which are considered aspects unique to the human mind.

Comments: 4 Pages.

Download: PDF

Submission history

[v1] 2015-09-07 18:59:16
[v2] 2015-09-08 18:58:24

Unique-IP document downloads: 66 times

Vixra.org is a pre-print repository rather than a journal. Articles hosted may not yet have been verified by peer-review and should be treated as preliminary. In particular, anything that appears to include financial or legal advice or proposed medical treatments should be treated with due caution. Vixra.org will not be responsible for any consequences of actions that result from any form of use of any documents on this website.

Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.

comments powered by Disqus