Large-Scale Deep Learning for Content-Based Recommendation Systems

近年来，人们提出了深度学习算法，以使机器学习系统朝着发现多层次表示的方向发展。 近年来，他们在许多传统的人工智能应用中取得了重要的经验性成功，如计算机视觉和自然语言处理。该提案重点关注将其扩展到用户建模的广泛应用领域所涉及的挑战，特别关注推荐系统。推荐系统在在线用户界面、数据驱动营销、计算广告和视频游戏中变得重要。这里针对的挑战涉及深度学习算法的规模和效率，以及将推荐系统扩展到个性化内容生成的更一般任务的机会。

Deep Learning algorithms have been proposed in recent years to move machine learning systems towards the discovery of multiple levels of representation. They have had important empirical successes in recent years in a number of traditional AI applications such as computer vision and natural language processing. This proposal focuses on the challenges involved in extending their reach to the broad application area of user modeling, with a particular focus on recommendation systems. Recommendation systems have become important in online user interfaces, data-driven marketing, computational advertising and video games. The challenges targeted here regard the scale and efficiency of deep learning algorithms and the opportunity of extending recommendation systems to the more general task of personalized content generation.