>Business >How to get the most from ML literature and courses

How to get the most from ML literature and courses

There are a ton of machine learning literature and courses available and there is an emergent trend towards free university courses and ebooks. With such brilliant resources available it can feel like you’re drowning. There is so much content that it can prevent you from beginning or making progression. 

In this blog post we wish to share tips for self study that enable you to touch a resource once, obtain everything we believe we can extract from that resource and integrate it into our own knowledge base so we can leverage it going forward.  

The knowledge base strategy and tactics we leverage were developed at a university while finishing a myriad of degrees and higher degrees. Even though beneficial in a formal setting. They are in self study settings where there is additional pressure to be effective with your time and resources.  

Knowledge Base Strategy 

Take up a knowledge base strategy. A knowledge base is a base of material you have developed that assists your memory. You do not have to recall the details of everything you have researched, just what you have studied and where your notes are situated. 

The knowledge base strategy needs that you develop a directory on your computer or in the cloud. In that directory you develop a document for every course and book you finish or read that details what you got to know from that material and where to go to obtain additional detail. For bigger courses you might require to develop a subdirectory and a document for every week or module of the course and every assignment. The particular breakdown of directory structure and documents is up to your discretion. 

The principle is that you touch upon every resource a single time and process it comprehensively so that you can make ideal use of it again in the future as you require to. This usually means reading or consuming it and taking obvious notes. 

Guide to Machine Learning Courses 

Courses contain video, text, audio, and in person components. Your goal is to internalize the content you are exposed to from the course, correlate it to the knowledge you already have and encapsulate that comprehension in summarized format. 

Courses can also have extra reading and resources, in addition to examinations, tests, tutorials, exercises and assignments. Within reason, these resources must additionally be processed in a similar manner. They are all just utilities developed to assist you in absorbing the material and must exploit them to your advantage through integration of the knowledge they expose into your knowledge base. 

Listed below are five tips for obtaining the most from a machine learning course. 

Regular Schedule 

Develop a routine so that you can finish the course work at a consistent and regular schedule. For instance, one hour a day every alternate day or eight hours every Saturday. Block the time out in advance and make effective usage of that time.  

Routine can get you a long way. When the scheduled time comes around you do not wish to have to think about what you have to do within that time block. You must allocate tasks prior in advance to prevent any overwhelm or over-thinking. At the conclusion of every time block, assign the tasks for the next time block so they are up and waiting for you. 

Your time estimations will be incorrect by a factor of 2 or 4. Consider at least doubling the time you schedule or halving the workload you pre-assign to those times. 

Make lecture notes 

Courses are demarcated into modules or weeks or some schedule of the sort. Every interval will possess materials that you will be expected to internalize. It is very probable that the course will have a summarization of the material in the shape of notes or lecture slides. 

Develop your own summarization of the material for the interval from the ground up. This will need that you take notes from the material that you consume in the interval then boil it down into a logical structure for summarization. This might be the same structure leveraged to present the material to you, but it is more often than not different. It can be different as you will have connected the material to knowledge you already possess and can see a clearer way to put forth the material to your own mind than the way in which it was put forth. 

Aim for approximately a single page summary for every module or week of material. This will demonstrate variance with the amount of content, but the goal is to go about summarizing the content to the critical concepts and a trail that links those ideas. Any detail you require on those ideas will be available in the original material like lecture videos or required reading. 

Study Group 

Create a study group of peers and advisors. You require people at your own level that are going through the material at the same speed so that you can discuss the content and make sure that you comprehend it. Learning in groups can be more effective than learning in isolation, particularly if you can pair up with others that are more sophisticated or quicker learners than yourself.  

Your research group can also consist of people that have already finished the course that might not be going through the material with you in concert but have concurred to answer the odd question or talk about the odd concept over with you on Skype. 

Leverage the group to test out your understandings. A strategy we liked to leverage was to volunteer to explain ideas even we didn’t comprehend them well enough. It is recommended to share all notes with the group for commentary and we would develop practice tests from the material for the group and others to finish. Contemplate about creative ways of obtaining the most from the group setting while also providing back to assist others. 

Finish Assignments Early 

Courses nearly always have tools to evaluate your acquisition and internalization of the material. This is in the shape of things such as tests, examinations, guides, assignments, essays, and exercises.  

It is particularly critical to take up assignments for self-study online and courses as there is usually reduced opportunity for learning retention to be directly evaluated. 

Observe that these materials at the start of the course and schedule to finish them early on review your outcomes on an ongoing basis. For instance, if there is a formal examination or evaluation, locate sample exams or evaluate from prior iterations of the course that you can practice on, even prior to exposure to all of the material.  

If there exists an assignment, develop a minimal version of your submission early and include to it as you get to know more and receive exposure to mre of the content of the course. You might have to actively seek it or even ask for prior year’s assignment from the institutor. Leverage this as an opportunity to obtain as much extra content as possible, they are all aids you can leverage to learn the material quicker and more comprehensively.  

This is a potent strategy as it will force you to seek out materials so respond to questions and assignment info prior to your readiness. There is something with regards to the pressures of this process that makes the things you find out really stick. When you receive exposure to the material in the course your understandings will be filled out and refined past the point of a starter. 

Group assignments are typically leveraged in university as an opportunity to push myself and finish the deliverables myself and then teach the content to the remainder of the group or as a baseline for the group to extend themselves further. 

Finish Extra Reading 

There is almost extra reading for you to finish. Handle this with vigour and leverage the procedures detailed below for internalizing this content into your knowledge base.  

Extra reading is usually in the variant of book chapters, papers and websites. Consider looking into your own additional reading to supplement the content and adhere to the same processes to internalize the material. 

Guide to machine learning books 

Machine learning books leveraged to mean a grouping of textbooks leveraged for postgraduate research. The domain has altered a ton in the previous few years and there is now an excellent range of applied books for starters that consist of assignments and practical tips and projects. 

It is recommended to begin with the applied books and shifting into the textbooks as your comprehension of the domain matures and your thirst for a deeper comprehension motivate you. 

You can look through books quicker than you can go through courses as they are discrete and typically consist of lesser material. If you cannot process a book swiftly is very probably a sign that you are not ready for it. Exploit this indicator and look out for resources to bridge the gap. For instance, if you are lost on a mathematical treatment of a strategy for an extended period, it is very probable a sign that you require to process some materials on the general mathematical strategy prior to tackling the particular treatment in the book. 

You can leverage books for reference to handle subject particular material just in time. The strategy in this section are not solely for this use case. Like the strategies for process of a course above, the strategies in this section are for process of a book in linear fashion from beginning to finish. 

Listed are five strategies for process of a machine learning book and integrating the material into your knowledge consciousness. 

Touch it once 

It is better to process every stage, every chapter, and the overbook slower than you can read it organically so you can obtain the advantage of only touching it once. 

Take your time and read through the content meticulously to make sure that you comprehend it and so that you can internalize it adequately that you don’t have to reread it. This might imply stopping if you are feeling fatigued or waiting till a quiet time so that you can provide the material a relevant treatment.  

Obviously, you can go through the book again, but your future references to the content ought to be for fine grained detail and not the general principles, as those should already be in your knowledge repository. 

Active Reading 

Active reading is independent from passive reading in that you are jotting down notes. You are an active participant in the material as it unfolds in opposition to passively consuming it with no action. 

Active reading is needed to make sure you obtain the most from the book so that you only have to touch it once. 

Chapter Summary 

Develop a one-page summary of every chapter after you have completed reading it. 

It is common practice to take comprehensive notes while reading but just the highlights of those notes become aspect of the chapter summary. We like the idea of one-pagers, the constrained scope forces concise writing.  

Book Summary 

After you have finished summarization of all the chapters in the book, develop a one-page summarization of the complete book. This must consist of a brief summarization of the critical material in the book, probably touching on every chapter. 

It is best practice to include a line or two about the motivation of the authors to make the book, or the positioning. This assists in framing the book in your mind’s eye and in your knowledge base when you require to refer back to it. 

Further Reading 

Machine learning is a technical topic and nearly all literature on it will contain references or further content for reading. Meticulously review the references and note each that you would like to read in follow-up and a note with regards to why. 

Integrate this list with your processed materials for the book. They can be useful to look back to in the future if you are seeking for something to read or when you are seeking to delve deeper on a specific strategy or technique. 


The concept of keeping and maintaining a personal knowledge base is potent if you are seeking to accelerate your education in machine learning, or any technical domain for that matter. 

Listed below are some additional general tips that you might find useful. 

  • Paper Archive: You will ultimately end up locating and going through a ton of papers on machine learning. This consists of conference articles, journal papers, dissertations and technical reports. Gather these materials, their details and links to these materials so that you can reference them again quickly. 
  • Active consumption: Leverage the active reading process to blog posts and research papers in the same fashion you would process the chapter of a book. If material is trash – more often than not it can be, stop reading and note this in your knowledge base.  
  • Algorithm descriptions: Maintain a corpus of algorithm descriptions. Leverage a template to formally detail an algorithm leveraging a handful (5-10) critical questions about the algorithm. Maintain the algorithms you have detailed leveraging the template together. You will very swiftly develop your own algorithm encyclopedia. 


In this article, we have gone through a technique for developing a knowledge base for efficient self education on the domain of machine learning. We then analysed five strategies that can be leveraged to add on top of this know-how from finishing online courses and reading courses. 

This is a complete process and a lot of these methods were produced for rapid learning as a consultant or as portion of formalized research programs where this is what we did all day. Pick and select the elements of this knowledge base strategy you wish and stick to them. 

In a short duration you will swiftly be able to tell a good book from bad or a good paper from bad. Additionally, you will intimately understudy your own learning style and your requirement for more formalized training regimes will reduce as you will be capable of designing a training program for your self to execute.  

Add Comment