Intelligence Assessment for Early-Stage Software Systems Aimed at Human-Level, Roughly Human-Like AGI
نویسنده
چکیده
One of the many difficult issues arising in the course of research on human-level AGI is that of “evaluation and metrics” – i.e., AGI intelligence testing. It’s not so hard to tell when you’ve achieved human-level AGI -though there is some subtlety here, which I’ll discuss below. However, assessing the quality of incremental progress toward human-level AGI is a much subtler matter. In this essay I’ll present some thoughts on this issue, culminating in a couple specific proposals:
منابع مشابه
AGI Preschool: A Framework for Evaluating Early-Stage Human-like AGIs
A class of environments for teaching and evaluating AGI systems is described, modeled loosely on preschools used for teaching human children and intended specifically for early-stage systems aimed at approaching human-level, human-inspired AGI through a process resembling human developmental psychology. Some particulars regarding implementation of environments in this class in online virtual wo...
متن کاملIntroduction: What Is the Matter Here?
Artificial General Intelligence (AGI), roughly speaking, refers to AI research and development in which “intelligence” is understood as a general-purpose capability, not restricted to any narrow collection of problems or domains, and including the ability to broadly generalize to fundamentally new areas [4]. The precise definition of AGI is part of the subject matter of the AGI field, and diffe...
متن کاملThe Role of Specialized Intelligent Body-System Networks in Guiding General-Purpose Cognition
Human cognition is portrayed as involving a highly flexible, self-organizing ”cognitive network”, closely coupled with a number of more specific intelligent ”body-system networks” – e.g. those associated with the perceptual and motor systems, the heart, the digestive system, the liver, and the immune and endocrine systems, all of which have been shown to have their own adaptive intelligence. Th...
متن کاملThe LIDA Model as a Foundational Architecture for AGI
Artificial intelligence (AI) initially aimed at creating “thinking machines,” that is, computer systems having human level general intelligence. However, AI research has until recently focused on creating intelligent, but highly domain-specific, systems. Currently, researchers are again undertaking the original challenge of creating AI systems (agents) capable of human-level intelligence, or “a...
متن کاملReport on the First Conference on Artificial General Intelligence (AGI-08)
main focus was on the creation of computer programs displaying intelligence at the human level and ultimately beyond. This proved harder than anticipated, and over the following decades most research activity in the field shifted toward the creation of computer programs solving specific sorts of problems that humans find to require intelligence. This more task-focused “narrow AI” work that has ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2008