** There are risks in employment, so you should be cautious about changing careers. * * * * * *

People kept coming up to me, asking me questions about changing careers. I work in machine learning, and some people don’t major in computer science and want to cross industries. Some people are already in the computer field, but what they do doesn’t seem so promising to them, so they want to look for new segments.

Indeed, for some industries, employment is difficult, such as history, civil engineering, business administration in recent years are not particularly good employment. These students, who have few opportunities in their major, want to find new job opportunities in different fields under the pressure of supporting their families. There are quite a few successful cases around me.

I have a high school classmate who majored in civil and architectural engineering as an undergraduate. At that time, I chose this industry according to the advice of my family. After going to school, I found that the civil construction industry is not very prosperous, and the demand for talents is gradually reduced. Since 2014, the energy of the computer industry has been on a spurt, so he began to get into computers. When he took the postgraduate entrance examination, although his ability was not enough to directly cross the computer major, he went to a 985 mile civil engineering laboratory that valued computer application. At that time, deep learning image processing was gradually emerging in the civil construction industry, and he began to introduce deep learning into the laboratory to deal with some tasks of defect detection in construction, which did a good job.

In the past two years, in order to constantly update his practical skills, he has participated in many competitions related to computer vision. A few days ago, when I was helping to publicize a contest, I found that he won the first prize. I am very impressed by this progress. Now, the teacher in the laboratory also trusts him. After he transferred to a doctor’s degree, he was asked to do the professional application of relevant algorithms with some junior and junior teachers.

This should be a successful career change. It should be noted that he did not go directly to related fields. Instead, they solve the problem by taking a roundabout approach, introducing relevant computer technology into their own domain. These results are the best evidence to convince laboratory leaders.

Of course, not all people have such determination and perseverance, nor do they have better opportunities. Before making a decision, you still need to think carefully about whether you are really interested in the industry you want to enter and whether it is really profitable.

In fact, machine learning has only been around for a few years. A senior student graduated with a master’s degree in 2011. At that time, he was one of the few graduates majoring in machine learning. There were very few jobs for him, so he was worried that he could not find a job. So, he worked hard toward the development direction, looking for a c++ engineer position. Later, my resume was attracted by an engineer of a large factory. I finally found a student who was engaged in machine learning, just like finding a treasure. It doesn’t matter if the algorithm is not deep, because the basic knowledge is not good.

This situation began to improve until 2014, when various enterprises began to pay attention to the importance of algorithms. Alibaba also launched the Tianchi Data Competition platform in this year, and many students had the opportunity to practice their skills through the competition. In 2016, machine learning and deep learning were really hot. Talents in related fields were in short supply, which led to a large number of students pouring into this field.

In the past two years, more and more people came in, and everyone began to say that it is difficult for people who do algorithms to find jobs. For those of you who change careers, it’s not one or two points more difficult.

Everyone has his own ambition, if the professional really do not like, envy the salary of the computer field, you can through their own efforts, turn over. But if this professional employment can also, oneself also feel still ok, the proposal does not easily go through the muddy water of the algorithm, thousands of troops and horses over the canoe, difficulty can be imagined.

For those of you who are interdisciplinary, you have to be careful, but for those of you who are already in computer science and want to go into this little branch of algorithms, what do you have to think about?

My attitude is basically to discourage, unless you are really motivated. Because most students want to enter the reason is that algorithms can make a lot of money (debatable), do not like the current job (avoidance psychology), many friends are doing algorithms (herd psychology). Very few people tell me I like algorithms, I like computer vision, I like data mining, I like natural language processing.

Sure, making a lot of money is a good goal. Do you have relevant papers, have you had a good internship or work experience, or have you ever done well in an important algorithm competition? If not, it is impossible for big factories to consider such people, after all, the screening cost is too high, resume can not pass. Small factories will give some opportunities if they can’t hire people.

A feasible and reliable route is that if you want to switch to Machine Learning algorithms, you should first read books, such as li Hang’s statistical Learning method, Zhou Zhihua’s Machine Learning, and PRML (Pattern Recognition and Machine Learning), which focuses on the basics. Still have a few actual combat courses on the network, some still quite depend on. Once you have the foundation, you can join some competitions, such as Kaggle, Tianchi, and many other emerging competition platforms in China. The data on these platforms are good, and some of them even provide computing resources. After that, I will work in a small company for two years and then go to a big company.

We often say look for opportunity, but everyone can see, is not called opportunity, that is the phenomenon, the phenomenon of conformity. Everyone comes, the profit space will be divided up very small, the cost of entry will rise sharply. Before you make a decision, think hard about whether you can really get a piece of the pie. Again, employment is risky, and changing careers requires caution.

Note: the menu of the official account includes an AI cheat sheet, which is very suitable for learning on the commute.

Highlights from the past2019Machine learning Online Manual Deep Learning online Manual AI Basic Download (Part I) note: To join our wechat group or QQ group, please reply "add group" to join knowledge planet (4500+ user ID:92416895), please reply to knowledge PlanetCopy the code

Like articles, click Looking at the