1. Extensive and accuracy of difference and iteration:
Extensive Reading: within a month, read 50 papers, rough read, understand multi-dimensional time-series signal, which research problem, challenges, research groups, research people.
After Accuracy: learn how details of how to build the model, build a bridge between what the model and reality.
Most begin extensive reading, intensive reading and then again extensive reading, intensive reading, loop iteration.
Although some do not understand, but can be studied on the basis of do not understand, and then ask again.
2. Strategic and tactical questions:
My strategy is, I do not know what to play?
Meng brother is a tactical question, hit the United States, how to play?
3. Reality VS model
When you propose a model, at the time of the reality modeling with a model, there will be between the gap, how to build a bridge between the gap, which belongs to scientific discovery.
4. On the research problem, do not listen to Liu, and to control yourself, when you are done extensive reading, you can put a halo around him, he needs the results of the current situation is that you give him a little achievement on it, on the issue and mentors to get along, you can abstract strategic discussions, not influenced by emotional factors.
5. Extensive range: Correlation Time Series: pattern discovery, prediction, visualization and the like. There are those who understand these cattle do, what group, what has been done. Figure papers and networks can also see, but very time-consuming, requires a lot of accumulation in order to produce results.
6. Read KDD and vis article.
Supplementary question:
ToG> PAMI = IJCV;
computer vision > image processing;
IJCV and PAMI are the top journals in computer vision.
ToG is considered the best in computer graphics