"Reprint" Learning Guide for machine learning beginners (experience sharing)

Learning Guide for machine learning beginners (experience sharing)

2013-09-21 14:47

I computer research two, the professional direction of natural language processing, individuals interested in machine learning, so began to learn. So, this guy is a rookie ... It is because of their own is a rookie, so realize the hardships of self-study machine learning, so here to share a personal experience, hope to be helpful to the beginner.
Some of the introduction of machine learning here does not do a detailed introduction, interested students can go to Wikipedia. Just go straight to the chase.
1, to Coursera Andrew Ng's "machine learning", complete all the work, it is best to get full marks. This is a fairly introductory course, and the teacher is an expert in the field of machine learning, and is one of the hottest experts in deep learning at the moment. You can go to Google for more information about the teacher. I started the course by registering half of the course, so I didn't get a high score at the end. But the real harvest is very big, so I look at the paper at the time relaxed a lot. The teacher is speaking in layman's terms, not worrying too much about math. And the work is also very suitable for beginners, are well-designed program framework, there is a job guide, according to the work guide to fill in the completed part of the line. This course is over, you can basically start a simple application of a variety of machine learning technology.
2. Find a project or find a slightly more specific book about machine learning. began to delve into the study. I am at this stage, looking for an open source project, but because of personal time and energy is limited, there is no time to complete. But still find a book, Hangyuan Li's "Statistical learning method". Because the book is written from the perspective of natural language processing, it is very helpful to me. It has been passed over, and the simplest model-the perception machine-has been implemented. is preparing to enter the implementation of the next model and carefully study the contents of the book. After learning and realizing the perceptual machine model to the lab students to do a report on the perceptual machine model, at this time I found that machine learning far more than I see so simple, application is only one hand, to understand thoroughly, but also to understand the story behind the model. I confess that all the questions raised in the report stimulated me! So enter 3 (personally feel can and 2 at the same time, of course, need time and energy enough).
3, the cultivation of internal strength. Say oneself found light application not, still have to understand its behind the east after, want to ask someone to consult, where to learn advanced strength? No, there's a pillow on the NAP. A group of machine learning people are fine on the microblog wound up, I also like to see their wound up. So I picked up the bag. Map a picture.

In other words, this is the machine learning Godfather level task Jordan recommended books, and all are internal secrets! I see after the collection, and then a notebook to download Ah! And then look at these English books, silent tears, it seems that graduate stage and sister paper is the fate of the wood!!! Because of my basic mathematics, I also found a child to optimize the entry of English materials (Introduction to optimization). Most of this information can be found in Baidu, I do not enumerate the list of resources. Let's take a picture. But I only downloaded the optimization section, because the rice had a stutter. The rest of the parts do not know Baidu can search. If the search is not available, you can go to Sina Weibo to find those who do machine learning (especially the more famous) to seek help, they will be very enthusiastic.
4, Walk here, the road has been relatively clear. The rest is to withstand the temptation of loneliness and sister paper, and the laboratory of computer and lighting through the Mid-Autumn Festival bar.

So much, I hope this rookie of a little experience to help you.

"Reprint" Learning Guide for machine learning beginners (experience sharing)