Toward a unified science of machine learning
نویسندگان
چکیده
منابع مشابه
Toward a Science of Machine Translation
In this paper, I suggest several ways to improve machine translation, based on the best practices of human translators, as described in Nida’s (1964) Toward a Science of Translating. I call this approach multi-pass machine translation (MPMT), as it crucially relies on processing the text more than once. It is similar to the opportunistic bricoleur approach of Gdaniec (1999) in that it sets out ...
متن کاملToward a Science of Learning Games
Reinforcement learning involves a tight coupling of reward-associated behavior and a type of learning that is very different from that promoted by education. However, the emerging understanding of its underlying processes may help derive principles for effective learning games that have, until now, been elusive. This article first reviews findings from cognitive neuroscience and psychology to p...
متن کاملGeophysics and nutritional science: toward a novel, unified paradigm.
This article discusses a few basic geophysical processes, which collectively indicate that several nutritionally adverse elements of current Western diets also yield environmentally harmful food consumption patterns. We address oceanic dead zones, which are at the confluence of oceanography, aquatic chemistry, and agronomy and which are a clear environmental problem, and agriculture's effects o...
متن کاملToward a Unified Model of Attention in Associative Learning
Two connectionist models of attention in associative learning, previously used to model human category learning, are shown to have special cases that are essentially equivalent to N. J. Mackintosh's (1975, Psychological Review, 82, 276 298) classic model of attention in animal learning. The models unify formulas for associative weight change with formulas for attentional change, under a common ...
متن کاملA Cognitive Science Based Machine Learning Architecture
In an attempt to illustrate the application of cognitive science principles to hard AI problems in machine learning we propose the LIDA technology, a cognitive science based architecture capable of more human-like learning. A LIDA based software agent or cognitive robot will be capable of three fundamental, continuously active, humanlike learning mechanisms: 1) perceptual learning, the learning...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Machine Learning
سال: 1989
ISSN: 0885-6125,1573-0565
DOI: 10.1007/bf00116834