Superior Artificial Intelligence
Abstract
The project goal is to substantially advance the state-of-art of artificial intelligence (AI). As a major approach to achieve this goal, we increase our understanding of the mechanisms that underlie biological intelligence. We avoid methods and systems that may appear intelligent, but are, in fact, hard coded and not intelligent. Many current systems give the appearance of intelligence; however, these systems cannot adapt to changing circumstances or truly interpret and make decisions based on input. Systems that can achieve flexible adaptation and decision-making represent the future of computer technology and are the target of this Superior AI project. This work builds energy aware neurocomputers to solve problems that cannot be addressed by today's AI technology. Successful completion of this work allows going beyond the state-of-the-art AI, which is represented by Deep Learning (DL). In the overall problem setting of DL, resource constraints are often ignored, or have secondary role. DL typically requires huge amount of data/ time/ parameters/ energy/ computational power, which are not readily available in various scenarios. Target applications include rapid response to emergency situations based on incomplete and disparate information, supporting graceful degradation in the case of physical damage or resource constraints, and real time speech recognition in noisy and cluttered background. In spite of the drastic cuts in project budget by the Program Manager from Year 2, and his demand of eliminating some of the tasks, reorganize the others, and completely deleting the final year of this 4-year project, several breakthrough results have been accomplished during Y1-3, as planned.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 21, 2020
- Accession Number
- AD1114098
Entities
People
- Robert Kozma
Organizations
- University of Massachusetts