- 74
- 29 332
Andrea Jones-Rooy
United States
Приєднався 8 жов 2015
Data science needs you! Come join the revolution!
Types of machine learning | 20 Data Science for Everyone @ NYU
This video provides a high-level overview of the three main types of machine learning (ML) as well as visual representations of how to think about the two key types of supervised machine learning: regression and classification. This video is great for all newcomers to data science looking to quickly understand how this part of the field is organized and build intuition for how it works.
00:27 Overview of three types of ML
00:38 Supervised ML
01:48 Unsupervised ML
03:21 Reinforcement learning
05:18 Creepy
05:44 Visualization of simple regression
08:39 Visualization of multiple regression
09:36 Visualization of binary (two-class) classification
10:08 Visualization of multiclass classification
10:31 Visualization of classification with multiple attributes
11:02 Example questions to practice translating the world into ML problems
This material is adapted from the NYU course Data Science for Everyone (DS4E). It corresponds to chapter 11.2 in the DS4E online course text: ajr348.github.io/ds4e_course/chapters/11_ml/02_supervised_ml.html
NYU Center for Data Science: cds.nyu.edu
Prof. Jones-Rooy: www.jonesrooy.com
Feel free to ask questions below or at jonesrooy on Instagram!
Data Science Needs You!
00:27 Overview of three types of ML
00:38 Supervised ML
01:48 Unsupervised ML
03:21 Reinforcement learning
05:18 Creepy
05:44 Visualization of simple regression
08:39 Visualization of multiple regression
09:36 Visualization of binary (two-class) classification
10:08 Visualization of multiclass classification
10:31 Visualization of classification with multiple attributes
11:02 Example questions to practice translating the world into ML problems
This material is adapted from the NYU course Data Science for Everyone (DS4E). It corresponds to chapter 11.2 in the DS4E online course text: ajr348.github.io/ds4e_course/chapters/11_ml/02_supervised_ml.html
NYU Center for Data Science: cds.nyu.edu
Prof. Jones-Rooy: www.jonesrooy.com
Feel free to ask questions below or at jonesrooy on Instagram!
Data Science Needs You!
Переглядів: 129
Відео
Statistics vs. machine learning | 19 Data Science for Everyone @ NYU
Переглядів 772 місяці тому
What is the difference between statistics and machine learning? Welcome to one of the biggest fights on the internet! Here's my take on their difference come at me! 00:55 What is the primary goal of statistics? Of ML? 01:47 Shout out to fidget spinners 02:08 Origin story: statistics 03:02 Origin story: machine learning 04:41 When to use stats vs. ML? This material is adapted from the NYU course...
Classification intuition for machine learning | 18 Data Science for Everyone @ NYU
Переглядів 492 місяці тому
There's a ton of excitement about machine learning, but what actually is it? This video is for newcomers to the world of data science particularly those with limited math and programming confidence to build intuition about how machine learning works. We cover the basics of the k-nearest neighbors classification algorithm. 00:16 Two types of supervised machine learning: regression and classifica...
Interpretation & limitations of OLS | 17 Data Science for Everyone @ NYU
Переглядів 982 місяці тому
Ordinary Least Squares is super powerful and super helpful but also dangerous if we get the interpretations wrong and/or violate some key assumptions. Join us to find out how to make the most of our OLS analyses. 00:03 Regrettable quantity of glitter on my face 01:05 Setting up the research question for this simple linear regression example 03:11 Some assumptions are already being violated! 03:...
Linear regression in human terms | 16 Data Science for Everyone @ NYU
Переглядів 4106 місяців тому
Linear regression is a foundational technique in statistics and is often among the first algorithms we teach in machine learning as well. But what is it? I break down Ordinary Least Squares (OLS) in human terms for all who are new to it (and those returning who want a refresher)! 1:05 What we are doing when we are doing regression the big picture! 05:11 Setting up an example 06:46 The linear re...
Correlation and Prediction | 15 Data Science for Everyone @ NYU
Переглядів 4536 місяців тому
We cover the fundamentals of correlation what it is, how to measure it, and what it can (and can't) tell us about our data. (By the way, just because "correlation is not causation" doesn't mean it's not super valuable; AND it also doesn't rule out causation!) 00:06 Four reasons we do data science 02:00 Association overview 02:20 Linear association 05:56 Measuring association with Pearson's corr...
Conditional Statements in Python | 14 Data Science for Everyone @ NYU
Переглядів 1918 місяців тому
New to programming? Wondering how conditional statements work in Python? This video is for you! 00:32 Conditional statements overview 01:52 If statements 02:31 If ... else statements 03:07 If ... elif ... else statements 04:10 Control flow statements This material is adapted from the NYU course Data Science for Everyone (DS4E). This video is number 14 in the course series, for those following i...
Loops in Python | 13 Data Science for Everyone @ NYU
Переглядів 1968 місяців тому
New to programming and want to know how loops work in Python? This video is for you! P.S. Come for the loops, stay for the Britney Spears reference! 00:22 Loops overview 01:23 For loops 04:03 Ranges in loops 05:56 Ranges on their own 06:31 While loops 09:31 Nested loops 11:25 Infinite loops This material is adapted from the NYU course Data Science for Everyone (DS4E). This video corresponds to ...
User-defined functions in Python | 12 Data Science for Everyone @ NYU
Переглядів 5758 місяців тому
This video provides an introduction to creating user-defined functions in Python! User-defined functions are super important and you'll see them again and again as you continue in data science (or programming for any reason, really). I hope this video helps you get comfortable using and experimenting with this technique! Don't forget to practice on your own alongside the video and with examples...
Programming pep talk! | *Bonus!* Data Science for Everyone @ NYU
Переглядів 1108 місяців тому
Do you want to learn how to code but are intimidated by it? Have you tried to learn to code and gotten frustrated because the first few things you learn are relatively easy, and then all of a sudden you have no idea what's going on? If so, this video is for you! One of the hardest things about coding for newcomers is that the learning curve gets really steep really fast. But I promise you that ...
Common errors in data | 11 Data Science for Everyone @ NYU
Переглядів 1718 місяців тому
This video covers four common reasons a dataset we are working with might be wrong, and what to do about them. To many of us conflate data with Truth, but data is a momentary, subjective snapshot of the world. This doesn't mean we throw away data we don't like. It means we must critically evaluate ALL data put in front of us, carefully consider how any errors might affect our inferences, and th...
Turning the world into data | 10 Data Science for Everyone @ NYU
Переглядів 2058 місяців тому
Measurement is the art and science of turning the world into data, it is wildly important, and tragically overlooked in the world of data science. We tend to get excited about fancy methods and programming skills, but being able to translate something we care about thoughtfully and accurately into data in the first place is crucial for making predictions and discoveries. This also happens to be...
Finding & working with data | 09 Data Science for Everyone @ NYU
Переглядів 2578 місяців тому
Congratulations! You've decided to study something! You take to the internet to find some data to start working with and you find ... not much. This video covers something that's both relatively simple but also really important and often overlooked when we think about data science what kind of data you're likely to run into out there, how to make sense of it, and how to tell whether it's any go...
Descriptive statistics | 08 Data Science for Everyone @ NYU
Переглядів 2959 місяців тому
Down with the tyranny of means! This video is for anyone who wants to do more with the data they have without having to learn programming or machine learning or any of that! Of course, I hope you will (and believe you can!) learn those things, but there's tons of room to improve our analyses between simply only calculating the mean for everything and needing to know what a neural net is. I prov...
Key terms in statistics | 07 Data Science for Everyone @ NYU
Переглядів 3939 місяців тому
Think of this video as a glossary of foundational terms you'd learn in any statistics class. Unlike with programming, most of my students have *some* statistics background by the time they get to me, so this video is more to highlight common terms, make sure we're all on the same page in terms of how we use them and what they mean, and set us up for more advanced work quickly. 00:22 Populations...
Python fundamentals for data science | 06 Data Science for Everyone @ NYU
Переглядів 3249 місяців тому
Python fundamentals for data science | 06 Data Science for Everyone @ NYU
Getting started programming in Python | 05 Data Science for Everyone @ NYU
Переглядів 6329 місяців тому
Getting started programming in Python | 05 Data Science for Everyone @ NYU
Causality in data science | 04 Data Science for Everyone @ NYU
Переглядів 6339 місяців тому
Causality in data science | 04 Data Science for Everyone @ NYU
Thinking like a scientist | 03 Data Science for Everyone @ NYU
Переглядів 6029 місяців тому
Thinking like a scientist | 03 Data Science for Everyone @ NYU
The power of the scientific method | 02 Data Science for Everyone @ NYU
Переглядів 9229 місяців тому
The power of the scientific method | 02 Data Science for Everyone @ NYU
Introduction to Data Science | 01 Data Science for Everyone @ NYU
Переглядів 3,7 тис.9 місяців тому
Introduction to Data Science | 01 Data Science for Everyone @ NYU
2.10: How to get out of your own way, with Gerrit Jones-Rooy!
Переглядів 212Рік тому
2.10: How to get out of your own way, with Gerrit Jones-Rooy!
2.9: The science comedy revolution, with Kasha Patel!
Переглядів 233Рік тому
2.9: The science comedy revolution, with Kasha Patel!
2.8: The power (and fun!) of collaboration, with Jay Novella
Переглядів 139Рік тому
2.8: The power (and fun!) of collaboration, with Jay Novella
2.7: Stay authentic, d*mn it! with Lisa Schwartz
Переглядів 170Рік тому
2.7: Stay authentic, d*mn it! with Lisa Schwartz
2.6: Do work that leaves a mark, with Chris Duffy
Переглядів 70Рік тому
2.6: Do work that leaves a mark, with Chris Duffy
2.5: How to create your own path, with Natalia Reagan
Переглядів 107Рік тому
2.5: How to create your own path, with Natalia Reagan
2.4: How to make a big life change, with Stephanie Humphrey
Переглядів 87Рік тому
2.4: How to make a big life change, with Stephanie Humphrey
Bonus ep. 5: Film & college critics unite!
Переглядів 312 роки тому
Bonus ep. 5: Film & college critics unite!
2.3: How to stay active in politics, with Dean Obeidallah
Переглядів 692 роки тому
2.3: How to stay active in politics, with Dean Obeidallah
I really want to pursue Data Science at NYU but my GPA is 2.3 HND STATISTICS ma I qualify for bachelor degree in data science
Was the answer to question at 4:02, maybe about having a hypothesis, and testing it/ falling to prove null hypothesis or it is because of inferring causality ?
at 7:25, truly world's famously "Breaking" scientist working in "Chemistry". Great one professor
The "thumb up" is happening, because you are using a Mac, and there is a built-in "Reactions" (gesture recognition) feature, enabled when you are using the camera. The "Reactions" feature creates playful graphical animated reactions (thumb up/down, disco, fireworks etc). You can control/enable/disable it by clicking on the green camera symbol when camera is active.
❤
NYU grad student here and this video did help!!!
It isn't easy being Dr Brian Wecht. To be so intelligent, so debonair, so deadly in seven forms of martial arts. To be so admired and feared by the public. To be so desired by men and women alike. A strong man, a smart man. A ninja and a gentleman that rocks your juicy testicles.
Heaton sent me, great overview and intro, looking forward to going through the course!
Aw, amazing! Thanks for being here -- and hi, Heaton!
Please make more videos! I have the exam for this on Monday and the past two videos helped me a lot! The ones before helped my midterm!
Ummmm this was really good!
"Is the variance of y constant with a change in x? Or does it increase or decrease as x changes?" I remembered knowing that once upon a time i knew that was important ... But I had forgotten; thanks for reminding me.
I'm glad it was helpful! It's an easy thing to overlook for sure! Thanks for watching :).
Thanks
Thank you for watching :)!
Great videos!!Please make more they help me so much in class this semester
"In the year 2000..."
hey really liked the video, but being an editor i can confidently say there's a room for improvement like titles are not discoverable and content could be made more engaging postproduction, if you feel there's something that we can about, then i open to reading your reply)
Wow wonderful video. All the videos on your channel are fantastic. But your video views are lower. Because your optimization is very poor. Your optimization needs to improve as soon as possible.,.,.
Thanks, Dr. Jones! I think this is episode 13 not 12
Haha, you know, I just noticed that as well! Thank you for catching it :)! Fixed!
that's great
thanks for watching :) !
Great presentation!
I’m a little disconcerted that what was once such an important cultural touchstone like the X-Files has become so obscure. 😨
Great foundation and definations this is really different aspect to think about data , and surprisingly life! Thanks Dr. 😍
Thanks
Thank you for being here!
Appreciate your effort Dr. Jones 🙏, on fire for the next videos
Aw, thanks for the kind words, and most of all for watching :)!
Thanks for your explain
Thanks so much for watching :)!
Thanks Doc... great and concise introduction. However, you went too fast here. Sorry but I'm not native English speaker
Hi there -- thanks for letting me know! I do for sure need to work on talking more slowly, and will do my best to do so!
Great! It took a little Googling to find a Python page that looked like the one you were using, but I found it and now I've bookmarked it. I just read your notes and I see you gave links to Python notebooks. A little history ... You mentioned R. If you don't know, R is the open-source version of S, which was developed at Bell Labs in the '70s. (Like the C language, they named it using just one letter.) I was in the same statistics department as John Chambers and Rick Becker. We had a large statistical subroutine library, and they developed S as a front end to the library, saving us from data management and tediously coding each analysis in Fortran. In 1976 I was off site doing an internship at AT&T, needed statistical tools, and so was their first "user community." One thing I remember was asking for a macro facility, which they created in a few days. They, of course, deserve all the credit. 🙂
Ah, sorry about the run-around with Googling! I should have made it more clear in the opening of the video itself that there would be some links. I actually had been thinking maybe I'd put together a video of actually setting up a Python interface -- nothing fancy, basically whatever steps you followed as you Googled, haha -- and now you've motivated me to do so! So thank you for that :). (Let's see if I can pull it off anytime soon of course, haha.) As for the history -- wow! This is amazing -- I did not know R's history traces back to Bell Labs, AND I've always been fascinated by Bell Labs, and I'm BEYOND impressed you played a role in all of this! When is your memoir coming out? I also insist you deserve some credit :). Gosh, you should come teach my class!! The history of programming is fascinating and I don't know nearly enough about it, myself. How cool! Thanks for sharing this :).
@@jonesrooy By the way, S was chosen as the letter for "statistics". Bell Labs' lawyers were real fussy about licensing S, so at some point, an outside group got permission to create R, the open source version, and R being a little less that S. By now I believe R is much more extensive than S ever was. This is about all the history of programming I know. 🙂 Thank you for your enthusiasm!
great video... just wanna know if there's any point in time where we gon get into hands on.?... excited for the course outline.
Thanks for being here :)! Yes -- we start coding in the next video, so get your typing fingers ready :)! I'm antsy to dive in, too!
Thank you for your great efforts!
Thank you for your kind words and for being here :)!
Great video, Thank you!
Thank you for watching :)!
<3
Very beautiful session 🎉
Thanks for joining!
<3
Ok I’m gonna be a data scientist and global citizen.
haha, yes, please!
Love it!! As a statistician, skeptic, and fan of the SGU, I believe this is just what the world needs! (And personally, I started programming with machine language and Fortran, moved up to visual basic, and now I'm looking forward to being introduced to Python.)
Hey! Thanks so much for the nice message! Always exciting to meet a fellow, well, all of the above -- stats, skepticism, SGU, all rule!! -- and thanks for watching! My prediction is you'll find Python much easier than your other coding (and especially since you have such extensive experience, too!), but you tell me :).
The elements song by Tom Lehrer is the original science comedy act
😻 p̴r̴o̴m̴o̴s̴m̴
ᎮᏒᎧᎷᎧᏕᎷ 😑
Thanks for having me!
This is great. Wish there was more!
This guy has no idea what he’s talking abiut
hahaha -- but he says it with such authority he must be right :-)!!
👍 love this episode.
Thanks, Travis! Thanks for listening :). I am a huge Heaton & Jeremiah fan. Glad you enjoyed!
It is fun to listen to this first…. and then come back to watch the video. There is so much that is filled in with the picture! Love it.
every time Noodling comes up, my heart grows one size
FIRST!
Personally, I'd love to do a science podcast with Brian, talking about the latest developments in the fields of science and engineering. I'd especially love to hear his thoughts on: --Black holes (and assuming it is possible eventually) how fast you would have to go to escape one --Whether or not he thinks that the supposed warp bubbles that NASA scientist Dr. Harold G. "Sonny" White claimed to have recently discovered are evidence that is actually possible to build a fully functional warp drive --How long it will be before we can prove that super symmetry exists (after all, scientists proved that Neptune existed using math, long before it was actually discovered, in the same way super symmetry is shown to exist in math, but has yet to be discovered using current technology)
I saw the screen shot, and my immediate thought was, "He is literally the perfect first guest for this show."
Haha, amazing! I'm so honored he agreed to do it :). And I so agree. He's amazing!
Is there a raw RSS feed (as opposed to Spotify or iTunes channel) for the audio portion?
Yes! Will grab for you :). Thank you for asking for this, and for your interest! Stand by!
Hi again! Just added it to the show notes, and also it's here: feeds.captivate.fm/majoringineverything/ (is this what you're looking for? I barely know what I'm doing, in case that's not obvious, haha). Thanks again for listening!
Beautiful. Bring the huge biases!
“To swim around in the middle of the arts and sciences” well shit if that’s not the slice of universe I didn’t know I wanted
hooray!! well, I'll see you in the deep end, then, my friend :)!
this is a show that i could have used 20 years ago. instead, i'll use it now! thanks AJR!
Haha SAME HERE!!! I'm actually making it in case someone invents time travel.