How to Learn Good
The past week I mostly took a break from the hardcore machine learning grind and instead spent my time going through the Zero To Mastery "Learning How to Learn" course (https://zerotomastery.io/courses/learning-how-to-learn/). I figured that while it might cost some time investment up front, making sure that I was learning efficiently and in the right ways would pay hefty dividends in the future.
While some of the course was a review of things I already knew, there were a fair few bits that I hadn't come across before. Overall I found the course very valuable and I'm already implementing some of the techniques in my routines.
What follows is an outline of what the course covers along with some brief notes for each section. Basically, it's a reproduction of the notes I took during the course. I'm writing it out in blog form to both cement my understanding of the concepts as well as share the material with anybody who might want to up their learning game.
The course is organized into 5 sections:
Principles
Lies
Pillars
Science
Techniques
I'll go through each of the sections, list the important points, as well as include a few notes about each. And please note, there's quite a bit of overlap and repetition among the ideas, which is kind of the point, as one of the keys to really make things stick is repetition, and hearing the same ideas again and again in different contexts.
Principles
Obstacles are a good thing - they force you to get out of your comfort zone, which is where growth occurs. And if it was easy, everybody would be doing it, and the skill you're trying to build wouldn't be valuable.
Compound Learning - we're in this for the long run. Aim for small improvements every day rather than cramming all at once. Huge gains will be seen in time. A riff on compound interest.
Failures Don't Count - You could've crashed and burned on 500 job interviews in a row. Doesn't matter one bit if you crush the 501st one. Nobody knows or cares about your past failures - all they see is the person you are now. Use your failures to help you grow.
Choice vs Chore - you really don't HAVE to do anything. Anything you do is a choice, and you have the freedom to make that choice. Start framing your productive tasks as something you choose to do, not something you have to do - because that's exactly what they are. Life is very much like an MMORPG in this regard.
Pareto Principle - The initial 80% growth in a skill can usually be achieved quite fast by focusing on the critical 20% that matters. Ask yourself, "Is this the best use of my time?"
Skill Stacking - instead of trying to be the absolute best in the world at a skill, which is incredibly difficult, combine skills that work well together to create your unique niche that you're top 1% in. Be pretty good at a few useful skills that make you one of a kind when combined.
Curiosity - aim to maintain a child-like curiosity and open-mindedness about the world.
Value Learning above all - eg, choose a job based on how much you'll learn rather than the salary.
Lies
These are things that have been pushed in recent years as philosophies to live by, but have largely been debunked.
Follow Your Passion - you don't need passion; you need creativity, control, and impact to enjoy your work. And the way you get those is to be So Good They Can't Ignore You (Newport). To be sure, an interest and meaningful connection to what you're doing isn't a bad thing; but die-hard passion is hardly a prerequisite to an extremely satisfying work life. Cast aside the passion mindset: "What can the world offer me?". Instead, adopt the craftsman's mindset: "What can I offer the world?". Aim to be useful.
You can avoid Risk/Discomfort/Nervousness - you can't. Not if you want to succeed. The only way to avoid uncomfortable situations is to be a complete non-entity, and be satisfied with living and dying entirely within the comfort zone. If that's you, great. More power to you. But if you're reading this then it's almost certainly not the case. Instead, see the nervousness as a challenge for you to overcome. If it helps, make overcoming the discomfort/nerves your goal, rather than the actual outcome of the interview/networking event/date proposal/whatever.
Trust this One Person - Seek out multiple opinions/perspectives and combine them. Most successful learners/masters understand there's pros and cons to everything and not one set way of doing things. Be wary of people who claim that their way is right way and disparage other viewpoints.
The 10,000 Hour Rule - popularized by Malcolm Gladwell in his book Outliers, this myth has been thoroughly debunked. Simply the time spent doing an activity doesn't have much to do with mastery - in fact, we all do lots of things and actually get worse as time goes on - driving, for example. A far better indicator of high level performance is time spent on deliberate practice, which we'll get to shortly.
Pillars
Everything is a Game - IE, everything can be trained. Adopt a growth mindset, lose the fixed mindset. Fixed mindset says: "I'm not smart enough to do this". Growth mindset says: "I'm not smart enough to do this...YET." Big difference.
Feynman Technique - to really understand a topic, you need to be able to explain it to a child in simple terms. A child is always asking the question, "But why?". Ask yourself this question about a topic you're learning and see if you can answer. This forces you to get crystal clear on the fundamentals. Speak in simple terms, not jargon - jargon is often used by people to make themselves seem smarter than they are.
Trunk-based knowledge - start at the fundamentals - the WHY? of the topic. Lots of people want to start and focus on the leaves - the latest fads/technologies/frameworks. But the leaves will fall off, to be replaced by others soon enough; the foundation of a healthy tree is a strong trunk and roots. Master the "first principles" - IE the roots, trunk, and big branches, before you get into the details/leaves.
Efficiency Trumps Grit - Working hard is important, but making sure you're studying the right things in the right way is more so.
Science
This section is a list of good learning principles and techniques that have scientific evidence behind them.
Focus vs Diffuse mode - your brain has two main modes of operation - focus and diffuse. As its name suggests, focus mode is when you're heavily focused on one thing - eg, taking an exam. Diffuse mode is the opposite, when you're not particularly focused on anything - eg, when you're taking a walk and/or daydreaming. Both are important. You'll want to switch back and forth between the two for optimal learning. Spend time in focus mode until you get tired, then take a break, then come back, and so on. Breaks are good - they are a sign of high performance (because you need a rest), not weakness.
Sleep - hardly needs to be said, but quality sleep is vital to learning at your best. During sleep your brain actually flushes out toxins from newly-opened channels between your brain cells.
Brain Training - your brain is like a muscle. At any moment, the synapses are either growing stronger or they're growing weaker. You'll want to exercise it - use it or lose it.
Feedback - do you want positive or negative feedback? Positive feedback tends to work best when someone is just starting out, but experts (at least, the ones who can check their ego) tend to prefer negative feedback because it specifically tells them where they can improve. Immediate feedback is best (hiring a coach/mentor can be a great investment).
Long & Short-term memory - for long-term retention, revisiting material each day for 20 days is far superior than revisiting it 20 times in one day.
Active vs Passive learning - Active learning is better. Passive learning results in the "illusion of confidence". What is active learning? Taking notes instead of just passively watching a video/reading an article. Testing yourself - ask "How can I summarize what I just learned in my own words?".
Intrinsic vs Extrinsic Motivation - Extrinsic motivation includes things like money, fame, or other rewards. Intrinsic is things like autonomy, mastery, and purpose. At the beginning, extrinsic can be useful to get you started. Later, intrinsic motivation takes over.
Goals - use the SMART system. Break your goals down into almost laughably-easy sub-steps to eliminate the fear of failure.
It Pays to Not be Busy - healthy, relaxing leisure time is very good for the brain. Breaks, naps, quality social time with people you enjoy are all good. Find a way to not be busy (delegate).
Chunking - break big, complex subjects down into "chunks", and seek to understand each chunk. Then, combine the chunks together to get the complete understanding. Not only does this make the task of learning a new subject less daunting, but practically it results in the "bottom up" (or Trunk-based) learning we talked about earlier.
How to Solve Problems - Knowing how to solve problems is the most in-demand skill and probably always will be. The reason you learn any other skill is really just to become better at this skill. This is what businesses pay their employees to do, and they pay a lot of money to people who do it well. At its most basic, even the star of your favorite sports team gets paid because they are an expert in this skill - they solve the problem of scoring more points for their team than the other team and entertaining the fans. Some problems are better solved in different ways. Some problems require focus mode, others require the creative thought and innovation that comes from diffuse mode.
Deliberate Practice - the learning process should be tough. Temporary frustrations when learning are a good thing. You want to be right on the edge of your limits/just outside of your comfort zone. The three components of deliberate practice are specific practice goals, intense focus, and immediate feedback. If you reach a point where you've automated your performance, you've stopped improving (and for most things, that's perfectly OK). Generally, the solution isn't to "try harder", but to "try differently".
Spaced Repetition - a little bit each day > cramming. Repetition builds the neural connections. The Anki flashcard program works on this principle. Also see this cool interactive article about creating a "Leitner Box": https://ncase.me/remember/
Be Adventurous - novelty creates new neural pathways. Monotony is bad for the brain. The world is a playground.
Have an Endpoint - EG, tell yourself "I'm going to stop work at 5pm no matter what".
Be Bored - being bored allows for the onset of diffuse/creative mode. Don't automatically reach your favorite app to scroll when you're bored as this doesn't let your mind wander. Constant distraction is bad.
Techniques
We've finally made it to the last section. These are the techniques built upon the principles, pillars, and science that we've just gone over. The idea here isn't to start doing all of them all at once, but rather pick and choose a few that resonate and seem like they'd be a good fit for your personal routines and study methods. If things get too routine (remember, monotony is bad), revisit this list and pick a couple new techniques to mix things up. Now on to the list.
Pomodoro - you've no doubt heard of this tomato-clock-based technique before. The standard version recommends doing 25 minutes of work followed by a 5 minute break and then repeating, but you can customize it to your preferences. Using this structure forces you to switch between focus and diffuse modes of thinking. You can also use the fairly short and easy time commitments to help overcome procrastination.
Chunk the subject - essentially Divide & Conquer. Break down big subjects into smaller, easier-to-digest chunks. Khan Academy is a great example of this. They break down complex math topics into approachable 6 - 10 minute videos.
Spaced Repetition - think riding a bike. You revisit something so much that your brain basically says, "Hey, this keeps coming up, so it must be important - let's remember it!". It then gets committed to long-term memory where you don't need to revisit it nearly as often to remember it, or ever. Ideally, use different techniques each time you revisit something. EG, test yourself one day, teach someone else the next, draw a diagram of the topic the next, and so on.
Deliberate Practice - What's one level up from what you were doing yesterday? As discussed above, aim to be just past your comfort zone. Feeling a little uncomfortable is a good thing.
Create a Roadmap - What path do you want to take to get to where you want to go? Create a curriculum. Focus on what will make you feel good about accomplishing (intrinsic motivation). Chunk the subject. Have a clear end goal and timeline in mind.
Interleaving - Mix up your learning. Variety is good. Multiple sources, mentors, perspectives, etc. Mix up your practice as well. Don't be a baseball batter who trains on just fastballs - train on all different types of pitches.
Einstellung - refers to a rigid mindset. You want to avoid this. Many breakthroughs are made by people relatively new to a field because they don't have Einstellung. Einstellung occurs because you get to level where you stop taking advice/feedback and aren't open to new ideas. The antidote to Einstellung is basically to adopt a beginner's mindset. Always be willing to be told your ideas are wrong, and un-learn ideas if necessary. Keep a growth mindset.
Importance of Community - be a part of a community when you learn. This helps keep you accountable and you can much more quickly identify blind spots through feedback from your peers.
Habits - create an easy ritual to help you slide into deliberate practice/focus mode. EG, build a habit of doing a light stretching routine before you move onto your heavy weightlifting workout. The stretching routine is easy to do, and even though the weightlifting routine is hard, you'll easily flow from one right into the other once you've made it a habit. One popular way to build a habit is the "Don't Break the Chain" method, popularized by Jerry Seinfeld.
Systems vs Goals - The goal shapes the process. Set the goal, and use the goal as a tool to determine the system to get there (IE, reverse engineer it). Then focus on the system rather than the goal. The system will become a habit.
The Power of the Senses - involve metaphors, analogies, and stories in your learning, not just the facts. The more memorable, the better. Invoke as many senses/emotions as possible. Seek out new experiences, and learn in places that stimulate you (IE, a desk right next to your bed in a room you rarely leave isn't great. A beautiful sunlit room next to a big window looking out on a garden is better, or a cafe in a bustling part of town that energizes you). Make your notes funny and memorable.
Pareto Principle - Figure out the critical 20%. Ask "is this the best use of my time?". Look at job postings and find the most common recurring items. Remove the noise.
Parkinson's Law - work expands to fill the time available for its completion. Set small goals/chunks with constrained timeframes (aka timeboxing).
Deep Work - deep work = deliberate focus. Set up a dedicated work space. Set an end time. Make sure you're rested and energized. The opposite is shallow work, which can be done without your full focus.
Stakes and Rewards - what are the consequences if you don't achieve your goal? Stikk is a site that will hold your money in escrow, which you forfeit if you don't achieve your goal. Small rewards are the perfect encouragement.
Concepts > Facts - Facts are easy to look up; but people who can understand and relate concepts are in high demand. Be a child - ask "why?" often.
Test Yourself - instead of just watching videos/reading passively, pause every once in a while to test yourself. Ask yourself "What did I just learn?" and "How can I summarize it in my own words?". Reflect on WHY you got something wrong, not just what.
The First 20 Hours - you rarely need to become master-level at a skill. Often, you can become "good enough" quickly, even just the first 20 hours. Create a roadmap, identify the critical 20%, chunk the topic, test yourself regularly, etc.
And that's it! This is the summary of ZTM's "Learning How to Learn" Course. If you made it to the end, you clearly know how to read good, and now you know how to learn good too!