Year: 2022

Wow. It’s been quite a few 6 months. From all the 2 am college essays to studying calculus during lunch and other periods, it’s been definitely the most transformative period of my life to date. 

I think personally I’ve definitely learned that stress can push me to work a lot harder than I thought was possible. Again, having to write college essays at 2 am due to unfortunate procrastination was never ideal, but something I’ve definitely learned from. With the largely retrospective nature of college essays, I find myself being more reflective and analytical than I ever was, something that has helped me appreciate the smaller, less spectacular things in life. 

Let’s start with multivariable calculus, a double-edged sword of being academically rigorous (in a good way) yet also mentally and physically taxing. From spending pretty much every morning in math help asking questions, I definitely have learned strategies to use when I’m struggling in college. 

Here is my really delightful multivariable study guide on just the formulas we have to know! And this is just half of it….

Posting on the blog site it looks a bit blurry… I guess you’ll have to take my word for it that there’s a lot here!

Then college applications, the other Beloved-like haunting of my senior year. Working at 2 am, taking power naps for an hour to keep working (and hoping that I don’t fall asleep), all helped me grow. 

Personally, I’ve learned not only how to study, but how to cope with friends, something that I never really did in the past with bad tests/ college rejections. I’ve also seen how consistent friendships are much better than friendships that are more on and off. Hanging out with a few people for the entire semester, through the highs and lows of the year I’ve learned so much not only in academics but also about them. 

You always hear people say how you should start college applications early, yet no one actually does and ends up telling the rising seniors to do that, and repeat. I find either strategy has its merit, while yes, working early and having time to refine and build off of your claims systematically and being able to conference with others is good, I believe (while not procrastinating) working closer to deadlines is not the worst thing. Most people do the latter, which sure, a week or two here or there will definitely be bad, but your writing isn’t diluted with the criticisms of 10 other people, it’s true to you. 

So my advice is, even if you have many applications due in a very short period of time, as you probably will, don’t sweat it, but make sure to pull some pretty late hours to study. 

As for school wise, I would say generally keep up the rigor of your previous year’s classes. Me, I was lucky that a lot of the harder classes I wanted to take conflicted with each other, and I had to find substitute classes that were a bit easier. With the study hall that comes when you take online PE, working during lunch, and generally finding time, I was able to get a lot of my homework done, so getting home I could start on college work pretty quickly. 

As I’ve mentioned before, I definitely feel as though college applications and AP lit have made me a more retrospective and reflective person, something I value for growth. I also learned to appreciate the jobs of writers, as sometimes writing even a paragraph can take an hour.  

What lessons would I take away with me to college…. That’s a good question. Finding a group of friends to study with and have fun with is important, probably having to stay up until very late hours at the library studying, etc. 

This semester is by far my hardest yet… and I’ve learned so much from it. Hopefully next semester I’ll be able to learn the values of a more chill senior year. 

That’s been me for this short yet intense chapter of life. 

-Max

Reading Algorithms to Live By: The Computer Science of Human Decisions by Brian Christian, it’s easy to get immersed in the many different applications, explanations, and more that Christian provides. One such topic I’d like to talk about more is the idea of caching, and how it shows that when organizing any collection of objects, oftentimes no system at all is more efficient than a certain sorting pattern.

You’ve probably heard of caching, I believe in years past the district installed a “Clear Cache” extension that totally did nothing, but it was a cool little trash bin that said it cleared your cache! Caching is the idea that computers store pieces of information or processes that need to be frequently used. Imagine, instead of a brightness button, if you wanted to change your computer screen you had to physically go to a terminal prompt and input your commands there, like below.

That would be super unnecessary and inconvenient! Caching helps to make computers run faster, having key processes saved to improve efficiency. 

The trickiest part of this idea is, with only a limited amount of memory that your computer can hold in this “cache”, what should we keep and what should we get rid of to maximize efficiency? 

It turns out, while there are many complex solutions, like caching reactively where your storage depends entirely on what’s being used at the present moment in time, the overall most efficient strategy is a last-in-first-used strategy. Especially with a computer, information most recently accessed has the highest chance to be called again. (For more caching examples, check out: https://docs.aws.amazon.com/whitepapers/latest/database-caching-strategies-using-redis/caching-patterns.html!)

How can we visualize this? Imagine doing your math homework, let’s say your processes of completing this task are writing with a pencil, fixing mistakes with your eraser, solving problems with your calculator, and correcting mistakes with a red pen. Let’s visualize each of these objects as different processes on your computer. 

Let’s say you’re working with your pencil, solving a problem. Then making a mistake, you quickly erase it. What will you pick up next? Most likely your pencil to continue the problem. Lets say, after solving that problem, the next one requires a calculator. After picking up the calculator and solving the problem, you write down your answer with your pencil. Realizing you forgot to add a value, you quickly adjust your calculations and fix your work. 

Here, we can see that you kept picking up the school supplies that you used last – this idea of the most recently used tool being the one most likely to be used again.   

So how does this apply to our lives? 

Imagine you are a librarian, sorting books. I love this picture below with the annotations and circling marks – it perfectly gets at my claim. How should you sort books? Should the big Shakespeare book go together with the Romeo and Juliet books, even though one is a biography and the other is a work of fiction? 

This problem is what librarians face in their day-to-day operations: how should they sort these books to be efficient for readers to search for? 

Yet applying Caching concepts, and the idea of last used as the most likely accessed process, it makes the most sense for librarians to never reshelve a book. Since we know people will want to use the most recently used process, in this case, a certain book, why should librarians ever sort these books back into the shelves where it’s harder for patrons to find? Not only does this take time out of their day, but it also wouldn’t be efficient for readers. 

Which is super strange! It seems obvious that we should sort items, and order the chaos. But here, the most efficient strategy for librarians is to have a “recently checked out” book display in the lobby of the library, and only reshelve the books that have been on this display for a certain amount of time, let’s say a week. 

This realization leads us to question if reshelving books is even needed, and if not why so many spend their time doing something inefficient for both the librarian and reader. 

These ideas are super interesting to explore: it shows how some jobs in our society, or even entire processes in our society, could be unnecessary and even inefficient for society. 

Overall, I hope this blog helped show how we can use algorithms to not only better our lives, but learn a thing or two from the computers that power our society. 

You might have seen the recent news about Open AI or more specifically DALL-E— an AI system that can generate an entire picture from a simple text command. Usually generated images look slightly unnatural – as predicted since this AI can’t physically see. Yet sometimes the images yield surprising results, below are some of my favorites. 

Inputting “Salmon in a river” generates salmon fillets in a river swimming like live fish, which is super strange! You could ask anyone on the planet to describe salmon in a river – and no one would think of raw fillets in the water. We can start to see how data pollution can affect generated images. Here, there have been so many pictures of raw/cooked salmon over the actual fish, so DALL-E can’t discern the difference between the two and chooses the more popular one. 

 

Here’s another interesting one: inputting “Chess match on a pizza”, not only do you get the physical pieces and somehow a chess board instead of cheese, but I’m especially impressed by the first image. DALL-E chose to include 2 people playing on this pizza chess board – which again shows how much input data affects this AI’s understanding of the game chess as a 1 on 1 board game rather than just pieces on a checkered board. 

One last cool example of DALL-E is when inputting “Future of City of Kigali”  DALL-E generates the image below:

For context, Kigali is the capital city of Rwanda, a country in central Africa. As things currently stand, they’re ranked roughly in the bottom 20% of all Nation’s Gross Domestic Products, a statistic used to show the

 economic growth and output of a nation. Yet DALL-E is pretty confident that this developing nation will one day become a futuristic city with an emphasis on greenery. 

All of these images show how far AI has come, especially in visualizing our world. Thinking about it, DALL-E has boiled down our complex 3d world into 1’s and 0’s that computers use, which seems a bit out there. 

As an avid reader of the New York Times, I was recently captivated by this article, “A.I. Wrote These Thanksgiving Recipes. Would You Make Them?” by Priya Krishna. Krishna used another AI, Open AI’s GPT-3, a program that takes input and can give a plethora of not only recipe ideas but full measurements and instructions. Some of the inputs for GPT-3 were basic background questions like describing your heritage, to which Krishna typed: 

“I am originally from Texas, and I grew up in an Indian American household. I love spicy flavors, Italian and Thai food, and desserts that are not too sweet. Some ingredients I frequently cook with are chaat masala, miso, soy sauce, herbs and tomato paste.”

Not only did this AI create full start-to-finish recipes, but also created fusions based on thanksgiving dishes and Krishna’s culture. One recipe was Pumpkin Spice Chaat, a variation of a traditional Indian dish. Not only did GPT-3 have a spark of inspiration to fuse two cultures together, but wrote introductory notes (usually tips and tricks written by the experienced chef to aid the home cook, or backgrounds on why they cooked the dish) in the voice and perspective of Krishna. Here is the full top-to-bottom recipe: 

Pumpkin Spice Chaat

Written by the A.I. system GPT-3.

I grew up in an Indian American household, and I love spicy flavors. I often do not enjoy traditional Thanksgiving food because it is too bland. I like recipes that have different kinds of flavors and textures, that balance savory with sweet and spicy and tangy. This recipe for pumpkin spice chaat is perfect for my tastes! It is made with pumpkin purée, chaat masala, ginger, cloves, nutmeg, black pepper, honey, lime juice, cilantro, mango, red onion and chickpeas. It is a delicious and easy snack or side dish that is perfect for Thanksgiving.

Sidenote from Max: This introduction sounds so authentic, the voice and reasoning behind this fusion dish is so clear!!! (End sidenote)

Ingredients:

1 can pumpkin purée

1 tablespoon chaat masala

1 teaspoon ground ginger

¼ teaspoon ground cloves

½ teaspoon ground cinnamon

½ teaspoon ground black pepper

1 tablespoon olive oil

1 tablespoon lime juice

2 tablespoons chopped cilantro

¼ cup chopped roasted peanuts

½ cup yogurt

Instructions:

3. In a bowl, mix together the pumpkin purée, chaat masala, ginger, cloves, cinnamon and black pepper.
4. Heat the oil in a pan over medium heat. Add the pumpkin mixture and cook for 5 minutes.
5. Add the lime juice and cilantro and cook for another minute.
6. Serve with the peanuts, yogurt and lime wedges.

(End Recipe)

Before I describe how the dish went when Krishna actually made it, let’s revisit DALL-E for a second. Krishna decided to go a step further, inputting “Pumpkin Spice Chaat” into the image generation platform. Below, one of the images is what DALL-E generated, and one of the images is what Krishna made as per GPT-3’s instructions in her kitchen. Which is the fake one?  

It turns out, the left one was generated by DALL-E. Shocking! The vivid colors, pumpkin pieces on the Chaat, dots of yogurt, not to mention the LITERAL PIECE OF PUMPKIN in the background, make the image look authentic. This trick of putting the raw ingredients in the background is something that every food blogger does – as I’m sure you’ve seen at one point or another.

Overall, Krishna rated the flavors of the Chaat as a “grassy-flavored mush”. While sure, the flavor aspects of these recipes are lacking, I’m thoroughly impressed by the depth of both recipe creation and image generation of these dishes. Maybe one day our recipes passed down from generation to generation will be indistinguishable from those generated in minutes by AI. 

 

Imagine, in an alternate universe known as Algorithmopolis, college applications work a little differently. Instead of applying to colleges that you want to get admitted to, colleges send admission letters to students they would want on campus. 

Which is great and all, no supplemental essays to write! Yet, in this parallel universe, let’s say these admission letters are very committal. Once you open your acceptance email, you have to either fully commit or fully reject that school on the spot, never receiving another chance to attend the school upon such rejection. 

You know you received 100 acceptance letters, but you have no idea which schools sent acceptances. How do you commit to the best school?

In the world of Computer Science, this is known as an optimal stopping problem. When is the optimal, or best, time to stop reading acceptances and commit to a school?

 

The classical example in the computer science field is quite similar, in a pool of 100 applicants to be your assistant, at what point should you stop interviewing and commit to the best applicant, given the fact that you must reject or accept an applicant on the spot?

 

This problem sounds downright impossible at first – you’ll either commit too early to a worse school or reject the best schools in search of something better, or have to settle for a worse school in the end. There are 100 letters you have to sift through, the best acceptance has an equal chance to be the first as it does to be the last (and everything in between). 

It turns out the answer is surprisingly simple, in fact, it’s a number: 37%. To describe further, you should spend a little over ⅓ of your time, or in our case acceptance letters, seeing what’s out there and noting down the best school you saw in this period. After this searching phase, anything in the remaining 63% that was better than the best school you’ve seen you should commit to. 

Let’s quantify: In our 100 letters, the best solution is to open the first 37, with the intent of rejecting them all. Again, there is no definitively best acceptance letter, you wouldn’t know if your #1 school sent you a letter, you can only compare two schools for the better one.

After looking at these 37 letters, let’s assume the “best” school you saw and had to reject was the University of Illinois at Urbana Champaign, not bad! The average person living in Algorithmopolis would be stressed, they just had to decide to reject or accept an offer to study at UIUC! But you know the Computer Science deities and mathematical odds are ever in your favor. 😉

With this in mind, you swallow your fears and trek on. After all, the odds of receiving a better acceptance letter are high. After 10 letters, nothing. 20- same thing. In the 30s, there are a couple, but nothing worth committing to. Nearing the 50 mark, opening and rejecting every school in your path, you stumble across a letter, not your #1 school but definitely a top 5. 

Not bad! The normal person would’ve probably committed to UIUC, but you had patience and a bit of luck. (Not to say UIUC is bad! Let’s just say it was your 6th favorite school). 

Now, since you’ve been accepted to a very good school, let me explain why the 37% approach is so effective. 

Using probability, let’s work from 1 acceptance letter to the 100 that your scholarly self received. With 1 letter, congrats! Since you (probably) want to go to college, accept the first and only letter! Celebrate with some cake. 

With 2 letters, you know the best offer for acceptance has a 50/50 chance of being the first letter you read. Why we exemplify the first letter is important, it shows what we miss out on with each subsequent letter we receive. 

Going to 3 letters, the chance of that first letter being the best one yet is ⅓, 4 being ¼, and so forth to 100. 

Let’s focus on receiving 3 letters before things get too messy. We can either accept the first letter, it is technically the best one we’ve seen yet. Or, hold out until the last letter, being forced to accept it. Both result in many cases where we have to reject the best offer. Our answers lie in the second letter. Applying our 37% rule (roughly), setting a baseline with the first letter and rejecting the offer, and then choosing to accept anything better that follows is surprisingly effective. 

The math is interesting here: this strategy of looking for ⅓ of the time, and committing to any school after, raises your odds of choosing the best acceptance letter to 50%. Look at that! You just turned this 3 acceptance letter problem into a 2 letter problem. (With 2 letters you’ll choose the best letter half of the time).

How do your odds rise to 50%, rather than the 33% of just randomly choosing a letter? With the schools, we can visualize them in the order of reading as 3-2-1, with 1 being the best offer. There are 6 possible scenarios here, and applying our rule of 37%, you’ll commit to the best school half of the time (3-2-1, 2-1-3, and 3-1-2), commit too fast to a worse school ⅙ of the time (1-2-3), and be forced to choose the third school ⅓ of the time (1-2-3, 1-3-2). 

Look at that! With a little bit of logic, computer science, and wit, you’ve gotten into the best college, given the terrible college application process in Algorithmopolis. Now to worry about moving in….

 

Inspired by “Algorithms to Live By” (Brian Christian) and Ted-Ex riddle videos.

This summer, I was fortunate enough to have the opportunity to not only get a free flight to San Francisco to attend one of my favorite types of events: a Hackathon. 

Being somewhat interested in computer programming freshman year, I attended a coding event, or a Hackathon, run by Naperville North’s Computer Science club. This event was eye-opening: coming from a nonsocial middle school background, hanging out with like-minded students, and doing something I enjoyed (coding and problem solving), not to mention the endless pop and snacks we were provided, I quickly loved this kind of social event. 

Fast forward to the pandemic, from competing in watered-down virtual hackathons to attending webinars, these Hackathons were very much incomparable to their in-person counterparts. 

To end that somber note! There is a national organization, Hack Club, that helps to organize these coding events and sponsors high schoolers to run Hackathons. This year’s summer, they organized “Assemble” – a riff off of Marvel’s Avengers’ catchphrase. Their goal was to “assemble” all high schoolers to their San Francisco HQ, and they even offered travel stipends!

Since I was free about two weeks before school would start, I eagerly signed up and hoped I would get a stipend. Much to my excitement, I was able to get the stipend! I remember that summer afternoon when I raced to my computer to find a $500 flight that was reasonable (this was our stipend).

I was really excited to have the opportunity to get flown out to San Francisco for the sole purpose of coding! This would be pretty much my first vacation ever since the pandemic started, and more especially my first time flying/traveling alone. 

I will admit, I spent much more time than I should’ve trying to make sure I wouldn’t get lost at O’hare Airport (and the San Francisco Airport). Looking at maps, and reading Slack (a communication platform) for Assemble, I was increasingly more excited to go. 

Whew! With all of that exposé, I’ll touch a bit about me actually traveling!

Waking up at 4 AM on a Friday morning, my dad and I rushed to the airport so I could catch my flight. Arriving at the airport, I knew the lines for security would be lengthy, and my experience was no different. With the shortage of TSA and airport employees, I anxiously waited in line, hoping I wouldn’t miss my flight. 

Getting through security – I was thrust into the chaotic and busy environment of a waking airport. Quick sidebar – in Chinese, there is a word for a really lively and happening place – 热闹(rè nào). What I find so great about this phrase is that, in English, rè means hot and nào means noisy – which almost perfectly describes the late summer chaotic terminal I had just stepped into). 

I have, of course, taken flights out of O’hare before. Like relearning a forgotten skill, I found myself walking somewhat familiar routes and slowly finding my gate. Shockingly (or really not so shockingly) it wasn’t that hard to follow signs around an airport I was familiar with. As I sat down and waited for my boarding group, I knew the hardest part was about to come – navigating SFO (San Francisco Airport). Not only was it an unfamiliar environment – but I had read on Slack that I had to take a tram inside of the airport, and I had to navigate through an unfamiliar place. (In fact, the organizers made a video on walking through the San Francisco airport, and it was 8 minutes long and cut in some parts). 

The flight itself was pretty uneventful – waking up at 4 AM meant that I would be spending my entire plane trip sleeping. 

I would like to take this opportunity to talk about this kid – let’s call him Window Seat Kid. Considering my boarding group was last, this meant the plane was pretty much full when I got onto the plane. 

My seat was the window seat, which happened to be occupied by Window Seat Kid as I found my seat. 1) I was too tired to argue too much and 2) the Window Seat kid refused to move. So I’m glad I pretty much slept for 4 hours straight! It made my newfound middle seat a bit better.

 

Landing in SFO, the unfamiliar environment was, to put it lightly, confusing. Terminal sizes twice as big as O’hare, travelers and employees alike yelling to each other, and leaving the dimly lit plane cabin into the bright California sun,  I was definitely thrown off. 

Then I had to find the train. 

(A preview for next week)