一位在MIT教数学的老师总结了十条经验,对理工科学生应该很有教益:
1.你能够做到每天七个小时坐在书桌前。
2.只有在学你觉得学不会的东西时才能学到东西。
3.总的来说,知其所以然比知其然重要得多。
4.在科学和工程方面,没有人能骗太久。
5.并不是天才才能做有创造性的工作。
6.你必须对自己高标准严要求。
7.世界变化很快,你最好选学那些坚实恒久的学科,少赶时髦。
8.你永远赶不上进度,别人也一样。
9.未来的计算机文化是正发生在你的身边,并不是计算机文化课上学的东西。
10.数学仍然是科学界的女皇。
存一些自己的Tips:http://t.zhizhihu.com/- 机器学习问题QA系统:http://mla.zhizhihu.com/
GRE201106TOEFL20110821大众点评网“我与点评有个约会”项目竞赛,上海DP总部答辩Web-Search Ranking with Initialized Gradient Boosted Regression Trees- The Elements of Statistical Learning
Support Vector Machine Active Learning with Applications to Text Classification- Parameter estimation for text analysis
Latent Topic Models: The Gritty Details- Finding scientific topics:http://www.pnas.org/cgi/
reprint/0307752101v1.pdf - Probabilistic Topic Models:http://cocosci.berkeley.edu/
tom/papers/SteyversGriffiths. pdf - introduction to topic modeling for the CACM
- On Optimization Methods for Deep Learning
Parsing Natural Scenes and Natural Language with Recursive Neural NetworksDistributed Gibbs Sampling of Latent Dirichlet Allocation: The Gritty Details- Structured Topic Models for Language
- Latent Dirichlet Co-Clustering
- Social Influence Analysis in Large-scale Networks
- Who Will Follow You Back? Reciprocal Relationship Prediction
- Conditional Random Fields: An Introduction
Computational Rationalization: The Inverse Equilibrium Problem:The Best Paper award of ICML 2011- read papers about graphical models
- Statistical Topic Models for Multi-Label Document Classification
- LSH Algorithm and Implementation (E2LSH)挺好用的一个近似快速的knn
- Relative Attributes ICCV2011 的Oral
- Applied Regression Analysis, A Research Tool, Second Edition
- Ensemble of Exemplar-SVMs for Object Detection and Beyond
- Curriculum Learning for Latent Structural SVM
- Tracking using Pixel-Wise Posteriors (ECCV08)
- Multiple Object Tracking using K-Shortest Paths Optimization
Demo:Human Counting in VideoHidden Markov ModelsMax Entropy Models- Sparse Coding and varieties
- Finding PhD or Job positions!!!!!!!!!


这是神马主题啊?名称?
知更鸟的
在留言一下
顶部小格子错位了啊
菜鸟一个,要向楼主学习。
回复中的踩和顶功能不错
这些效果好神奇啊 ,你是怎么做的啊 ?
truly right - i can never catch up my plan