Скачать или смотреть Patrick Hammer - Intelligence through reasoning

In this seminar, a real-time reasoning and learning system will be presented, and architectural challenges will be discussed. The human’s ability to reason has evolved in order to adapt to difficult situations and changes in the environment faster than current AI models allow for. Animals that reason effectively outsmart other species, and gain key survival advantages by doing so. Non-Axiomatic Reasoning can explain some of these cognitive abilities and provides a roadmap for cognitive enhancements based on psychological and neuroscientific insights. Furthermore, we will see how advancing this kind of reasoning and combining it with state-of-the-art Deep Learning models for perception allows applying the system in real environments, with the goal to enhance the autonomy of robots where human intervention is to be kept at a minimum. Finally, we will discuss how this will allow to more robustly perform tasks such as inspection and maintenance operations of city infrastructure such as power plants, which will ultimately lead to new technology for digitization which can help to solve environmental and societal problems.