Microsoft Research332 тыс
Следующее
Популярные
Опубликовано 17 апреля 2014, 19:07
Societies and governments around the world want to know how the biosphere is likely to change and what we can do to avoid or mitigate against it. Current models used to provide that information, though, are akin to miniature computer games: black boxes that convey almost no sense of confidence in their reliability. This problem can be addressed objectively by combining machine learning with process-based modeling to enable assessment and comparison of alternative model formulations to identify key sources of uncertainty and, ultimately, to enable probabilistic predictions of the likely consequences of climate and environmental change. Tackling problems of this scale, researchers have designed a solution using a new model-building platform Distribution Modeller and F# can be used to build and share data-constrained, process-based models and deliver their probabilistic predictions on demand through Azure.
Свежие видео
Случайные видео
Level Up Your Gaming Experience with BenQ MOBIUZ EX240N! 🎮 #GamingSetup #BenQGaming #NextLevelGaming