http://mp.weixin.qq.com/s?__biz=MjM5ODkzMzMwMQ==&mid=2650408190&idx=1&sn= f22adfb13fb14f8a220222355659913f
1. How to understand the status of NLP: see some tips for the latest doctoral dissertation
It may be a shortcut to look at the current status of an area and see the latest doctoral dissertation. For example, there are children's shoes asked how to understand the State-of-the-art of NLP, in fact, Stanford, Berkeley, CMU, JHU and other schools recently selected doctoral theses, the field of mainstream direction of the general situation can understand more than half. In addition @ William Wang 2013 recommended a streamlined list http://t.cn/RqQcgS6 and the following discussion throw worthy reference from the calculation of the text of the word
2. Course Video + Materials: Convolution neural network for visual recognition
The course video + Information : convolution neural network for visual recognition (Fei-fei Li & Andrej karpathy) "" Cs231n:convolutional Neural Networks for Visual recognition "http://vision. Stanford.edu/teaching/cs231n/index.html
. New Lesson (slides+video): http://vision.stanford.edu/teaching/cs231n/syllabus.html
notes:http://cs231n.github.io/
Cloud : HTTP://PAN.BAIDU.COM/S/1PKSTIVP from Love Coco-love life
3. Recent developments and practical tips for CNN (i)
Ph. D., CAs Melody, who has just uploaded a talk slides in VALSE2016, on recent developments and practical tips on CNN, on CNN's Progress and Caffe's common skills [for a tutorial on Caffe use], see links http://t.cn/ Rqqzadl from Tang Xu _shanghaitech
4. Alphago Related & ICLR16 Two & an asynchronous gradient descent optimization method for deep-reinforcement learning & UCL lecturer David Silver "Intensive Learning", ten lectures
Nature16 Alphago, ICLR16 Two (continuous action space, priority experience reuse), ICML16 a (deep reinforcement learning of the asynchronous Gradient descent optimization method), UCL lecturer David Silver "intensive learning" ten. Basic concept: Markov decision-making process dynamic programming predictive control. Actual combat: Valuation function Strategy gradient learning and planning development and utilization of game. 481 page Http://t.cn/RqQGlGG from eso9
Author's homepage: http://www0.cs.ucl.ac.uk/staff/d.silver/web/Home.html
5. Chris Olah, who received the Peter Thiel Scholarship, has several blogs about understanding and visualizing neural Networks: Calculus on Computational graphs:backpropagation,understanding LSTM Networks, visualizing Mnist:an exploration of dimensionality reduction,understanding convolutions
Address: http://colah.github.io/archive.html
6. Why should I focus on interest feeds
Published the headline: Why should I focus on the interest feed? On the one hand, Twitter and Instagram have announced a shift to interest feeds, and the trend is evident. In addition, the recommendation algorithm engineer to the traditional recommendation system to pay much attention to the interest of the feed this form of attention less, according to KK, it is more representative of an internet inevitable. Http://t.cn/RqQP0hl from Resyschina
Turn: convolutional neural Network for visual identity Course & recent progress and practical tips for CNN