Common Sense Reasoning in Autonomous Artificial Intelligent Agents Through Mobile Computing
Henderson, Angel Z.
MetadataShow full item record
CitationHenderson, Angel Z. 2020. Common Sense Reasoning in Autonomous Artificial Intelligent Agents Through Mobile Computing. Master's thesis, Harvard Extension School.
AbstractCommon sense reasoning is a critical branch of artificial intelligence that aims to provide intelligent agents the ability to learn, reason, plan, and apply logic in the same way that humans do. Developing intelligent agents capable of common-sense reasoning allows artificial intelligence to address some of the most complex problems that individuals and organizations alike deal with daily. However, in recent years the development of common-sense reasoning has not made as many strides when compared to natural language processing and computer vision due to the many challenges that arises in trying to develop a computer system that can perceive, learn, and adapt autonomously.
This thesis presents the design, analysis, and implementation of an autonomous artificial intelligent agent capable of planning, reasoning, rationalizing, communication, and learning through the use of common-sense reasoning. The foundation for the autonomous artificial intelligent agent is based around a hybrid computing architecture called the Artificial Cognitive Neural Framework which utilizes fuzzy systems, stacked artificial neural networks, and system modules to map out a cognitive process similar to the human cognitive process. The outcome of this thesis includes the development of a software mobile application containing an intelligent agent that applies artificial cognition to demonstrates an understanding of context, observations, utilization of past experiences, and manipulation of data to reach solutions. This intelligent agent contains artificial cognitive perceptors that captures a wide range of data types including text, image, video, audio, motion, and other forms of sensory data to extract knowledge, understanding, and experience through stacked artificial neural networks.
Lastly, Continuously Recombinant Neural Fiber Networks are utilized as the foundation for developing and mapping the memories of the intelligent agent, including short term memories, long term memories, sensory memories, and emotional memories. In addition, we aim to validate mobile computing as an effective platform for artificial intelligence research due to computational resources, advancements in machine learning capabilities, accessibility, and access to data.
Citable link to this pagehttps://nrs.harvard.edu/URN-3:HUL.INSTREPOS:37365032