Kozyrkov

Making Data UsefulHow to muscle up on data-related topics quickly(Feeling impatient? Scroll past the text and cat photo to get the learning paths!)Your author.First, may I say a huge thank you to all of you for encouraging me to write? I just noticed that here on Medium, my community of followers is 70% the size of Barack Obama’s. Whoa!I’m honored and humbled by all the love this amazing community has given me. I don’t know if it’s despite my being a cheerful weirdo or because of it, but thank you! And thank you for being…

A first look at two major AI releasesYour author. This isn’t the video, the video is lower down. Or here, if you insist.Here comes an AI unboxing video! These shiny new tools were released just over a week ago, so they’re fresh out of the oven. In the video, you’ll see me running my first ever Bard + GPT-4 side-by-side prompts. Below that, you’ll find something that started as the video transcript and quickly morphed into a feast of asides, edits, and snide comments. If that’s your cup of tea, enjoy!Link:…

The industry-wide neglect of data design and data quality (and what you can do about it)My favorite way of explaining the difference between data science and data engineering is this:If data science is “making data useful,” then data engineering is “making data usable.”These disciplines are so exciting that it’s easy to get ahead of ourselves and forget that before we can make data usable (let alone useful), we need to make data in the first place.But what about “making data” in the first place?The art of making good data…

The bias-variance tradeoff, part 1 of 3Should you read this article? If you understand all the words in the next section, then no. If you don’t care to understand them, then also no. If you want the bolded bits explained, then yes.“The bias-variance tradeoff” is a popular phrase you’ll hear in the context of ML/AI. If you’re a statistician, you might think it’s about summarizing this formula:MSE = Bias² + VarianceIt isn’t.Well, it’s loosely related, but the phrase actually refers to a practical recipe for how to pick a…

The bias-variance tradeoff, part 2 of 3In Part 1, we covered much of the basic terminology as well as a few key insights about the bias-variance formula (MSE = Bias² + Variance), including this misquote from Anna Karenina:All perfect models are alike, but each unhappy model can be unhappy in its own way.To make the most of this article, I suggest taking a look at Part 1 to make sure you’re well-situated to absorb this one.Under vs over… fitting. Image by the author.Let’s say you have a model that is as good as you’re…

The bias-variance tradeoff, part 3 of 3We covered a lot of ground in Part 1 and Part 2 of this series. Part 1 was the appetizer, where we covered some basics you’d need to know on your journey to understanding the bias-variance tradeoff. Part 2 was our hearty main course, where we devoured concepts like overfitting, underfitting, and regularization.It’s a very good idea to eat your veggies, so do head over to those earlier articles before continuing here, because Part 3 is dessert: the summary you’ve earned by following…

AI Art: Beauty And The BiasLet’s take a Midjourney around the worldWelcome! This isn’t so much a blog post as an exhibition made with Midjourney, showcasing the beauty and bias of AI art. My inspiration was wondering how AI systems reflect national identity, so I used the same prompt with different country names to generate some art and the results are fascinating!How do AI systems reflect national identity?I’ve pasted the 200 images below without comment, leaving it to you to draw your own conclusions.But first, some…

A primer on interpreting other people’s hypothesis testsIf you’re in the market for a new tongue-twister, try this paraphrase of a classic:“The difference between statistically significant and statistically non-significant is not necessarily significant.”Photo by Hello I'm Nik on UnsplashAs a recovering statistician, I have the pleasure of knowing many data experts and the displeasure of meeting a lot of posers. Though the people I bump into are hardly a random sample, I have noticed a correlation between throwing around…

Introduction to “good” statistical estimators and their properties“The bias-variance tradeoff” is a popular concept you’ll encounter in the context of ML/AI. In building up to making it intuitive, I figured I’d give the formula-lovers among you a chatty explanation of where this key equation comes from:MSE = Bias² + VarianceWell, this article isn’t only about proving this formula — that’s just a mean (heh) to an end. I’m using it as an excuse to give you a behind-the-scenes look into how and why statisticians manipulate…

Battling an embarrassing new alchemy for the digital eraThe data careers space has been disgustingly hyped up. Not overhyped (there is incredible value to be had from data), but more like “mishyped” — a lot of people are generating data buzz for all the wrong reasons.The right reasons to be excited relate to the old adage, knowledge is power: the power to improve your business, your work, your personal life, and the world around you. With all of the technological improvements in storing and processing the raw materials of…