Hi,
What is the most fascinating thing you've learned or discovered this month? Or perhaps these past few months?
Note: If you find this newsletter truncated by your email client, you can directly visit the published version here.
Imposter Syndrome and Abstractions
Lately, I've been having a lot of Imposter Syndrome. Whatever I do feels superficial. Perhaps, reading a lot of ML research papers (working on a lot of abstractions) has catalyzed that feeling, especially with a giant mess of mathematics.
Also, it has to do with a lot of interview processes over the work. I have to (re)think a lot of abstract ideas to assess the candidates.
Stepping outside of traditional "technical interviews", I've been trying to refactor the interview processes (at a personal level) by testing candidates’ problem-solving skills. It's okay to not know much about certain things. But, it's not a good sign to overvalue yourself based on what you had learned in X years. I see a lot of candidates diving into Machine Learning and Data Science without actually knowing depths of their personal projects. At least, you show know what worked and what didn't work, right?
I am not being judgemental of the candidates in any manner. But it has put me in a weird loop of assessing my own knowledge too...
If you know the problem, and possibly the solutions (and trade-offs) to some depths, it's fine for me to get impressed. There was one session with a senior candidate. I was reluctant to interview at first because you know...I don’t feel "eligible" enough to interview seniors. So, I approached the interview in a different way. Instead of jumping into technical questions, we discussed problems that his projects didn’t cover.
Anyway, out of the frustration accumulated from the interviews and research projects, I had posted this very-naive-and-superficial advice on data science over Linkedin. It blew up (in a good way). Now I regret having posted that because I am not in any position to spew out "wisdom" in the domain. I am too naive. For me, everything feels like black boxes!
This even applies to a lot of other things I do. Like playing guitar. Without substantial knowledge in music theory, I hit walls frequently, despite playing it for a long time. Probably, I lack creativity?
Aside: what started out as a draft titled “Abstraction and Intuitions”, ended up me being more contemplative about my learning processes (which is very inefficient, mostly procrastination overtakes). Now, I have complied with this notion of Curse of Knowledge:
The more you know something, the harder it becomes to get detached from it.
That is, it’s harder to let go of personal beliefs. (See: The Lesson to Unlearn)
So, instead of putting everything in one specific post, I have decided to segregate these topics for later; I want to get a clear understanding of the topics. (I’m still assessing many things.)
I believe this has more to do with one’s learning process than their ability to understand. Upon analyzing my own, I have diagnosed some vague factors that affect the process (in no way it’s a conclusion):
Feynman Effect
Sometimes I feel I have understood a topic. But I forget the intuitions later. This effect is known as Feynman Effect:
Listening to Feynman's explanations makes sense at the time of listening, but after a certain time, it's hard to recall what the topic was actually about.
Motivation and Intuitions
Motivation plays a vital role. This motivation is (probably) put in the shadows by our education system too. From a very early age, we aren't taught to have intuitions but engulf a large amount of information. Curiosities slowly attenuate when teachers put students’ questions on hold or leave out important concepts. In retrospect, teachers shape our curiosities, and hence our learning motivation.
For instance, during engineering days we weren't told how statistics or calculus are applied to real life. It's an irony that "implications" of an idea X isn't taught through engineering lens, despite being an engineering school. It’s more like “you have to rote-learn this equation, practice these questions, pass the exam…and get the degree".
Teachers and Mentors
Sometimes it's good to have a mentor or a teacher that really can guide us. Someone who inspires us to go in-depth on certain topics.
For me, while starting out to learn guitar, I wish someone had told me that it’s a bad habit to fiddle without any "proper" time signature. Although, I’ve been an auto-didact (someone who learns by himself), mentors could have brought in their experiences to make my learning process much efficient. Nevertheless, I’ve been re-iterating my habits. Cheers to all those tedious-long practice sessions without the internet. Haha.
Emotions
Emotions are intricate parts of our lives. They help us to connect with each other. With stories. With ideas. With ourselves.
If you are learning for the sake of just "accumulating things" (see: Collector’s Fallacy), without any connections to ideas, you are missing out on the point of learning. The collector’s mindset can only take you on an “expiring knowledge” train (I fall in that category). So, emotions are important to tie us to various topics.
For instance, people (re)watch Grant Sanderson videos (3Blue1Brown) because he presents explanations in a very "humane" way. He knows that many (theoretical) concepts are hard, yet beautiful. So, his videos follow certain stories with emotional values for the audience. (Again, I feel Feynman Effect also comes into play if you don’t apply the concepts anywhere.)
Analysis Paralysis
Instead of being meta, that is constantly thinking about thinking and learning, sometimes we have to act immediately. This is one of my “really” bad habits. I don’t take enough courage to implement things I have learned. Perhaps, because there is a lot of decision fatigue involved while thinking about the matter at hand. The same concept can be applied to many domains. Or the same problem might need a multitude of knowledge.
Mostly being meta is a good thing. You can assess the situation with care. However, the process itself compounds badly to next day, next week, next month. Just imagine the time you have wasted watching/reading “How to X”.
Mostly, during my “naive younger days”, I had a philosophy of “Learn by Doing”. You can get a hint of that through my projects: panim and playx. But now, things are a bit rough.
Anyway, you can find myself talking a lot (repetitively) about this topic in bits and pieces.
(figure: invisible mountains.)
#Reading
How to foster Shoshin
Christian Jarrett | 10 min
Shoshin is a Zen term meaning beginner's mind. Cultivating a philosophy of beginner's mind helps you to learn more and be open to new ideas.
This kinda reminded me of an old Zen saying:
"You cannot fill a cup that's already full."
The philosophy of Shoshin can also be related to the Curse of Knowledge (see above). We have a confirmation bias towards our own knowledge which, in hindsight, closes our minds towards new things. (Related: Idea Attenuation)
For instance, how often do you get offended when someone points out certain mistakes in your pre-established, dogmatic beliefs (and knowledge)?
Also, people tend to highly overestimate their knowledge. When asked to explain a familiar concept (say, how does toilet/zipper work), their initial reactions would have elements of over-confidence. But, eventually, they’d fail to give clear explanations. This not only happens to new learners but also in experts. I think it's a legit problem that stems from lacking "intuitions" on the topic. This phenomenon of overestimating personal knowledge is often termed as Illusion of Explanatory Depth.
So, what can be done?
Be open-minded. Be eager (and curious) to learn more.
Constantly assess your own knowledge. Know that the confirmation bias exists.
Use the Feynman Technique to learn. If you cannot explain something in simpler terms, you haven’t really understood it.
Wonder and ponder about things. Have empathy. Embed a sense of awe in you towards the world. (This hits close to home)
Have intellectual humility. Know that intelligence is a matter of perspective.
Learn in Public
Shawn Wang | 7 min
Learning in public is one of the best ways to challenge your learning process. The idea is to share publicly what you are learning:
Tweeting things (which I don't do because I quit social media...haha)
Maintaining blogs or newsletters (trying to do)
Making videos (vlogs might be good I guess)
Drawing (nope...I have no idea)
Teaching (helps a lot on self-assessment)
I think that teaching (or mentoring) can be an effective medium (possibly the best). It pushes you to go in-depth on a topic. It challenges your dogmatic beliefs and helps you explain the topics concisely and intuitively.
Another approach might be implementing things by yourself. It has helped me a lot. But then, there is so much one can do. Eventually, there comes a point when you need collaborations and mentors. Working alone just doesn’t scale your throughput of learning.
Being taught by someone who has had similar struggling experiences might help. Teachers shape your motivation but they are only your guide to the world. Don’t expect to get spoon-fed.
Just try to remember your teachers who inspired you. I bet there are only a few. I don’t even remember one…oh except Richard Feynman. :)
One thing I kinda disagree with the author is his idealistic view that audience size doesn’t matter. I think that’s a legitimate view. However, without any feedback mechanism, there's no way to know what you're doing (learning/thinking) is correct or not.
If you're learning things incorrectly you should correct them soon. Practice creates permanency, not perfection.
On that note: I believe one’s goal while learning in public should be to maximize their luck surface area, plus encourage serendipity.
The Hardest Part About Learning Hard Things
Scott Young | 5 min
Learning is frustrating. You get a lot of self-doubts. You start to think that it's hard for you, unlike other people. You discard the whole process and, often, get back to"I wish I didn't quit" loop.
This frustration and self-doubt stem from hitting a wall at some point while trying to learn certain things. The wall eventually becomes an invisible mountain. It's not just a single mountain. There are a lot of them.
I think the majority of it has to do with your beliefs about learning. Here, the author presents some plausible approaches that worked for him while trying to learn something:
Dive Straight In. Just do it.
Avoid Comparison. Comparison sucks. It makes you drowned in self-doubts. For me, this is one of the reasons I am refraining to get into social media.
Embrace the Frustration. Know that it exists. It makes you realize your shortcomings and opens a lot of loops to be refactored.
Everything I Learned about Programming by Going to Art School
Amy Dyer | 5 min
An inspiring read I'd say. Amy shares her experience of how art school had taught her to be creative not only in the domain of "art" but also in general problem-solving ones.
“Being creative” is not something that can be taught but experienced and enhanced.
I think once the creativity boosts up, it can be applied anywhere.
The main takeaway for me is -- despite teachers playing a key role in our early learning phase -- no teacher can fully know what you are supposed to do. They can only be the source of inspiration, not a full path we intend to walk on. Amy’s sentiments perfectly describe the situation:
“At the end of my first year of art school, I suddenly understood not only how far I had to go as an artist, but how far I had actually come as a programmer. I think I expected that one day a teacher would tell me something, and suddenly art would ‘click.’ ”
I think the more general lesson here is to push yourself. To achieve something -- even in problem solving or creative works -- you have to do it yourself.
The only secret lay in just doing the work.
#Watching
What Makes People Engage With Math
Grant Sanderson | TEDxBerkeley | 19 min
I am sure Grant is an inspiration to a lot of people (not just me). Here, he talks about how emotions and connections come into play while learning mathematics.
Generally, the beauty of mathematics is abstracted under symbols and technical jargon. They don’t seem “pretty”. But, learning math shouldn't be like that. Math should be about thinking deeply about simple things. Grant proves this through his simple, yet profound explanations (and animations).
Although math is like going to the gym, which only a few can endure, it is surprising how the majority of the audiences are attracted to his videos that are mostly theoretical and complex. This has something to do with the mystery and wonder of theoretical math. They are like reading fiction. You know you can't really apply fiction to real life. But they are means of warm escape; emotions that stick out when looking at mysteries of the world.
Also, one of the major parts of his explanations is storytelling. Instead of directly diving into the topic, his videos have plots. They make the audience curious.
Nevertheless, I often get back to this specific video to feel inspired.
Mary's Room
TED-Ed | 5 min
Mary's room, also known as Knowledge Argument, is a thought experiment that attempts to establish “knowledge” as subjective experience:
Mary has been living her entire life in a black and white room; she hasn't seen any color other than those black and whites.
However, she is a brilliant neuroscientist. She has been studying colors and waves for most of her life; wavelengths, frequencies...everything...
Now, when a red apple is introduced to her suddenly, will she gain any new information?
This is a difficult problem since she already knows about colors in general, does she get any new insights? Perhaps, there’s more to this experiment than this — a concept of Qualia — which is a subjective experience that cannot be described accurately. Like falling in love, seeing a color, feeling something...
Ludwig Wittgenstein presents this concept of subjective experience in his beetle-box thought experiment, also known as Private language argument (existential comics link haha).
A group of people is given unknown boxes labeled Beetle.
They don't know what “Beetle” means and aren't allowed to look into each other's boxes.
The question is: can anyone truly know what other boxes have?
This is a difficult philosophical problem. Since no one knows what “Beetle” is, each of them has their own subjective definition and descriptions of whatever there is in the box.
Related:: Is Your Red The Same as My Red?
Knowing vs Understanding
Richard Feynman | 6 min
If you know me personally, you can probably infer how much I talk about Richard Feynman. He has a profound impact on my life. Heck ya. I even had his “knowing vs understanding” sentiment at the beginning of my SOP. :)
This is one of the few videos I get back to when I hit a plethora of invisible mountains.
There's a big difference in knowing the name of something and actually knowing something about it.
Another Advanced Useless Machine
34 second
Watching this gives a good vibe. Haha. Not only the machine makes you feel it has a personality but also the engineering might have been meticulous.
Also, see this. Wholesome. :D
Fred Astaire's Famous Ceiling Dance (1951)
Reddit | 1 min 24 sec
This is amazing. Even if we know it's using a rotating set, it's fascinating. You’d want to watch this breakdown video too.
#Fragments
XKCD Comic: Dependency
This hits close to home. Haha.
Poetry: The Crunch - Charles Bukowski
Profound.
Music: Famous Blue Raincoat - Leonard Cohen
Timeless. Powerful. My favorite from Cohen. :)
Music: Desolation Row - Bob Dylan
This is profoundly moving (always). Dylan is a genius. Listening to him (and reading his lyrics) is like journeying through time and lives. This song gives me goosebumps. I guess that’s fine. :)
Also, I didn’t know about this story behind the recording of this song.
#Ending-Thoughts
This tweet by Conor resonated well.
Every day we swim in a large sea of information. I’d say it’s an avalanche. Deep rabbit holes here and there. But we rarely stop by and think deeply about them. We take information for granted. And that, I think, is a missed opportunity to learn many things. Taking notes is the first step in that I guess. At least that has been the case with me all these years. It has helped me a lot.
Signing off!
(I am sure I have left out a lot of things here like top-down/bottom-up learning…)
Nish
PS: Checkout this podcast - Sheldon Solomon: Death and Meaning | Lex Fridman Podcast 117. One of the best conversations I’ve listened to in Lex’s podcast. I didn’t know about Solomon. But now, I am definitely hooked in to learn more about him.