Deep Learning high-quality Communication Group

 

Recently, deep learning is very popular, but it is found that there are very few forums dedicated to DL discussion on the Internet, so communication is particularly inconvenient. I have just started to contact DL and cainiao. I hope to communicate with many like-minded friends. I have established a high quality QQ Group for deep learning, welcome to join. Group Number:209306058(Verification is required when you join the group. Please give me a simple explanation of your understanding of DL in a single sentence to avoid confusion. Haha, thank you !)

 

Required (required)

1. Stop uncivilized comments

2. Advertisement prohibited

3. Do not run the question

 

PILOT (refer to execution)

1.Content discussed in the group

To ensure the quality of communication within the group, please try to discuss algorithms based on deep learning, as well as related ml, DM, AI and other content... do not discuss other irrelevant content, even other IT technologies. After all, each group has its own subject. Thank you for your cooperation! If any uncivilized speech, personal attacks, advertisements, or other topics are found, the listener will be removed after warning! Please try again !! Any group member in the group is eligible to warn other group members who violate the rules. If any group member finds that the violation persists, contact the Group Administrator privately. Thank you for your cooperation!

　　

2. Regular cleaning

This group does not want to accommodate too many people. The Group will maintain a total number of people (such as 200, or fewer, depending on the situation), leaving the group members in the flow state, for those who never participate in the discussion, clean up the group within a certain period of time (for example, about 3 weeks), so that the Group remains active and healthy, and everyone will benefit from it.

 

3. straightforward

　　A common phenomenon is that asking questions doesn't directly throw a question, but uses the method of asking questions layer by layer, which is very lengthy. For example, if you want to ask a RBM question, first ask "Have you ever used RBM ?", Wait until someone answers the question and then throw the next sentence "what is the difference between RBM and DBN?" There is no need to ask first, just go to the topic.

 

4.Complete description

　A common phenomenon is that when discussing a topic, others know what you haven't said by default. For example, "many of the hidden layer features I have trained are close to full gray." The necessary content is omitted here: Which DL model did you use for training, RBM, autoencoder, or Sparse Coding or others? How many training samples are used, and what are the numbers of input-hidden-layer nodes? Try to describe the problem completely at a time to avoid being asked.

 

5. Avoid irrelevant content whenever possible

A common problem in a common group is that it is easy to spread over irrelevant topics. As a result, a topic often begins with a technology, and finally a hot topic on the Internet. There are too many names in the discussion, for example, "Great God" and "cainiao". During communication, everyone is equal and there is no need to do that. Ignore the red tape and talent habits and focus on the topic itself.

 

6.He must help himself first

Try not to throw the problem as soon as possible. Think about it first and then ask Google Baidu. As long as you think about the topic you want to discuss in the group, you can ask questions and discuss it, regardless of the difficulty of the problem. Welcome to discuss it.

 

7.Mostly used"High Technology"To solve the problem rather than using diligence

I often asked questions and asked questions. Should I reflect on how to improve the questioning skills, instead of copying and pasting the same question multiple times and failing to search for satisfactory results? Is it necessary to learn the searching skills first.

 

　　Form of discussion (proposal ):

For example, in terms of reading paper, for the purpose of in-depth discussion, any member of the group can propose the paper that they are interested in, publish it to the group, and set a time for discussion, ask other Members who want to read the article to contact you privately, and drag them to a temporary discussion group to participate in the in-depth discussion within the agreed time. However, the host (that is, the one who proposes to read the paper) should read the paper carefully in advance, prepare several insightful questions, and participate in in-depth and detailed discussions.

After the discussion, the host can publish the list of participants in the discussion to the group (so that the host is mistaken for being absent from the group for a long time and cannot participate in the group ), at the same time, we can make a summary of this discussion.

 

(If you have any comments or suggestions, contact the Group Administrator. Thank you)

This article has referenced the good quality communication group rule of Kinect + leap: http://www.kinectutorial.com /? P = 10