整理 | 周翔 作为深度学习的奠基人之一,Yann LeCun 的一言一行都颇受关注。近日,Yann LeCun 在 CCN 2017 (认知计算神经科学大会)上发表了主题为“How does the Brain learn so much so quickly”的演讲,内容详实,非常有启发性。 该演讲主要分为以下几个部分: 1)Obstacles to AI(通往 AI 的阻碍)
2)The Architecture of an Intelligent System(智能系统的架构) 3)Predictive Models In Questioning-Answering and Dialog(问答和对话中的预测模型) 4)Predictive Models with Uncertainty Adversarial Training(具有不确定性对抗训练的预测模型) 5)Video Prediction of semantic Segmentation(基于语义分割的视频预测) 6)Semi-Supervised Learning (adversarially)(半监督学习) 7)Fader Networks: Disentangling factors of variation to Parameterize the Image Manifold (adversarially) 8)How do we 'design' objective functions? We learn them 以下是AI科技大本营整理的完整PPT内容和视频播放地址: B站地址: https://www.bilibili.com/video/av15938572/ Youtube地址: https://www./watch?v=cWzi38-vDbE |
|