The MOST IMPORTANT topic in Machine Learning right now

[@ibrahimkhurshid129]

I don't know why youtube hides these awesome channels from my recommendation. Absolutely loved this content.

[@karolinasowinska]

Thank you! :)

[@duckmeat4674]

I've noticed a move away from blackbox models to clear ones instead, and I love it. A lot of companies will be moving towards this too

[@karolinasowinska]

Yep, definitely! :)

[@simonlauer9379]

That’s why it was so valuable to me to have graduated a quantitative economics degree. You learn so much about interpretability and causality. Great video :)

[@karolinasowinska]

Definifely, quantitative subjects are perfect for data science/machine learning!

[@sunshinenassima261]

You make want to learn this field, your way of talking about it is glamorous Love it

[@karolinasowinska]

Thanks so much, it's a really cool field indeed! :)

[@mryderoc]

Thanks for your great video. I've noticed you are not doing youtube video too much but I encourage you to do so because you're making a good stuff! Btw, I think your mic is not wreck, it's your necklace that made the noise! :D

[@karolinasowinska]

Thank you! I'm planning to create more soon! :)

[@bigdhav]

I just discovered your channel and wow! It's really refreshing and so full of candour. Thank you so much Karolina :).

[@karolinasowinska]

Thanks! ;)

[@calebmnb]

Really love the way you explain ML topics

[@griffmccoy5607]

When you combine the data sources you lose on explainability (though increase accuracy). For XAI better to have a separate model for each "kind" of data, then have them "vote" on the outcome. E.g. for a fraud detection algo we had one model whose features would pick up gibberish usernames, another whose features analyzed the IP address space, another that was trained on the user's behavior. Then we could see which of the "triggers" would go off. Deep learning does "aggregate" low level features into middle-range features (neurons), but these are not human-interpretable. You need to create your own interpretable middle-range features. You do lose on performance a bit this way. But you could have one model with all the data fused together that actually does the trading, and another set of simpler models that is trained on the output space of your complex model to "explain" it. Just like explaining human decision making, you have how you REALLY decided (instinct/intuition), and the linguistic rationale/justification that backs this up. This also comes up in court cases or medical diagnoses. Judges / radiologists can always find a justification for their decision, but is that how they really decided? No, it's all muddled up in their human brains.

[@siddharthpandey2118]

This was quite insightful. Thanks!!!

[@FarisSkt]

This was amazing ! Thanks for sharing your experience

[@sukeshseth989]

Not sure about the industry but I do face this issue with one of my friend, despite of having a great model he always fail to explain it to me, how the model is working, I'll surely share this with him, I hope he learns. Moreover, I've seen many people working with different toolkits like RASA & Kaldi but most of them don't know what's really happening inside, they just pick the data clean it & feed it to the model unsure about what's going to happen next. I really hope that people start focusing on model explainability.

[@karolinasowinska]

Very well said!

[@mattk3650]

Sharing this with all my associates

[@karolinasowinska]

Haha please do!

[@forecaster182]

I'm so excited I found this channel. I'm just learning ML because I want to do something similar to what you just described. Though I mostly expect my efforts to not yield any significant results on the stock market It's an exciting topic and will surely bring a lot of challenges which in turn will require me to learn a lot of new things. This is like a whole new pandoras box compared to my C/C++ dev dayjob 😅

[@karolinasowinska]

That's awesome! As you probably know, C/C++ devs are in price for many hedge funds that do algorithmic trading :)

[@forecaster182]

@@karolinasowinska I didn't know that actually. I'm in the industrial automation field right now :) Thanks for the info

[@empowercode]

Hey Karolina, nice to meet you! I just found your channel, love what you're doing! I like how clear and detailed your explanations are as well as the depth of knowledge you have surrounding the topic! Since I run a tech education channel as well, I love to see fellow Content Creators sharing, educating, and inspiring a large global audience. I wish you the best of luck on your KZbin Journey, can't wait to see you succeed! Your content really stands out and you've put so much thought into your videos, I applaud you on that! Cheers, take care, and keep up the great work ;)

[@karolinasowinska]

Thanks so much for your comment! I love connecting with fellow creators, I'll check out your channel! All the best:)

[@sheresh]

Thanks for sharing 🙂!

[@nathankomer8699]

Advanced ML approaches with explainable inner workings? I'm in! Thanks for sharing.

[@rodrigoamoedo8523]

As allways, grate job!

[@albertocatania6165]

May I ask you if you tried by yourself this system?

[@ChoogyNet]

Thanks. Machine learning algo are all pretty much same. Good to distinguish apple or orange, for now

[@karolinasowinska]

Some are quite similar for sure!

[@theshah2701]

Really amazing facts and empowering video Your heavy eyes are describing the hard wirk done or either the high weekends😅😅 overall really appreciable video and nature wish you grow more

[@karolinasowinska]

Hahaha yes, my eyes don't lie, I'm a little restless!

[@GodOfWar299]

I am master student in statistics and going to write my master project next semester with reinforcement learning with applications in the stock market :D Nice video. Did you try to give your algorithm real money to trade?

[@karolinasowinska]

I haven't haha. I might work on something like this soon though

[@roshan8853]

Any chance you could do a video on causal ML? I imagine it's relatively similar to explainable ML

[@karolinasowinska]

I'm sorry for my extremely late response. That's a good suggestion, I'll think about it. Thanks!

[@diahidvegi8536]

This was so insightful Karolina! And such fun story telling 😃 Thank you for sharing these insights! I better share a KoFi with you now 😃😊

[@karolinasowinska]

Thanks Dia! ❤

[@laurielounge]

Hi Karolina. Interesting story. One question remains unanswered to me. Totally get what they mean about black box modelling BUT - why do you think they agreed to meet with you? What do you think was in it for them?

[@karolinasowinska]

I told them I already had a job offer to join their company in a different department, which was true. So maybe they thought of me as a future colleague. Or maybe they thought they'll learn something interesting straight from academia!

[@mwansa430]

Nice nice👍👍👍. Do you also give career tips?

[@karolinasowinska]

Sometimes I do in the comments! Let me know if you've got any questions that I might help with. :)

[@mwansa430]

Thank you very much. Please answer when you're free😅. I was wondering what could be a good major for computer science. At my university there are two majors ... Software engineering and computer systems engineering. And if I chose software is it enough just to know how to code?

[@mwansa430]

I would like to go into software but I'm not very sure what the syllabus for my university is. From my research some universities have different syllabus for software. Some have cloud engineering some don't have.

[@mwansa430]

And they haven't posted it on the internet. I've also tried getting in contact with the administration. But one thing I have noticed is that all universities that have software engineering teach how to code.

[@deluded00]

Some really great explainer videos. A shame more people aren't watching or subscribing! How do you come up with the content for these? You should teach!

[@karolinasowinska]

Thanks a lot, that's really nice to hear! This is still a relatively new channel, I'm happy with how it's growing so far! ;) In terms of coming up with content, I draw on my previous experiences and think what could be interesting for others to watch :)

[@techprinvid338]

Hi karolina, Laptop 4gb ram is ok to learn machine learning or not

[@karolinasowinska]

I'd go for 8GB at least, you might find 4GB a little tight

[@techprinvid338]

@@karolinasowinska thank you

[@Kevinfdg]

Hi Karolina, I am graduating this semester with an undergrad degree in statistics and I was hoping to get some advice from you or anyone else who can advise me. I want to go to graduate school for Data Engineering. Does doing undergraduate research matter in the admissions process? I have zero research experience but I have had a lot of experience interning in tech (not in Data Engineering). If research is critical, then what can I do if I have already tried asking all of my professors and departments which denied me? What should i be doing to make myself more competitive? I feel like online courses wont carry as much weight as research in an admissions office. Any advice helps, thanks for reading.

[@karolinasowinska]

It's hard to say, because the answer depends on the specific graduate school you're applying to, and what their requirements are. In general, data engineering is not a research area, so this really shouldn't be an issue!

[@Kevinfdg]

@@karolinasowinska Thank you so much. Ive been looking into programs for data science and business analytics as well. Do you know if research is essential for applicants in these fields? One last question, do you have advice on what I should do in the meantime before I start my full time role in August to be a more complete applicant?

[@medotop330]

Can I implement ETL using Sqoop Hive or spark sql or should be implemented using Data stage or Talend and other vendors ETL tools ???....

[@khalidsultani6006]

cloud you please make a tutorial video about real work and how you use your programming skills🙏❤️🥰

[@karolinasowinska]

That's a very interesting suggestion! I'll try to do something like that! (but I might be a little bit limited as to what I'm allowed to show)

[@khalidsultani6006]

@@karolinasowinskayes we respect the privacy ,😉 thanks a lot.

[@ScantaniouslyCombust]

I'm absolutely not an expert, but I'm not sure if I agree with what they said about having a blackbox algorithm is as good as not having an algorithm at all. For example, if someone developed a blackbox algorithm that could predict that best course of treatment for any cancer with a high success rate, wouldn't that be useful? And speaking more philosophically, human brains are a blackbox in a way. Much of the brain is a mystery. (Most) Human beings can be pretty useful!

[@karolinasowinska]

I like what you said about human brain being a useful black box - that's very true! I'd have to reflect on this more deeply, but it's definitely an interesting question!

[@marvhartigan3677]

Thanks.

[@chouaibbenali5201]

Hey Karolina, great video,i'am a NLP student as you were and currently i'm working about Subjectivity,what advice would you give me for tacklng this subject ? I have free card for the problems i need to resolve but i have no idea how to start...

[@karolinasowinska]

Interesting one! Let me come back to you with some resources

[@denismerigold486]

it was interesting to listen to.

[@TimothyZhou0]

4:17 "let's print more money" lol facts

[@karolinasowinska]

Haha well... sad but true

[@DanielHelle-uj8dq]

What do you think about AutoML? Many people in the data science industry are afraid that it will automate away their jobs. I think it would make an interesting video as well. Cheers:)

[@karolinasowinska]

I might make a video about it, thanks! :)

[@RNSS0]

Thank you

[@dev5289]

stephen hawking liked the *someone from the future* part. Great video.

[@karolinasowinska]

Hahah thanks!

[@gabrielar.9047]

👏👏👏👏

[@medotop330]

Should I learn machine learning as data engineers ??

[@karolinasowinska]

You don't have to, but it won't hurt knowing it :)

[@arrymawlany3004]

can I ask to you in personal chat about the data engineer? I think your teaching technique is match for my learning 😣🥺

[@karolinasowinska]

Sure, feel free to DM me on Instagram, I hope I'll be able to answer your question!

[@Nancivlog]

Kindly make a video on basic project tutorial.. from scratch plz

[@karolinasowinska]

I'll see what I can do! :)

[@Nancivlog]

@@karolinasowinska thank u.. I am a Beginner..from India .....it would be really helpful, if u could teach a simple project tutorial from scratch

[@moustafaaymanx]

keep it up

[@karolinasowinska]

Thanks, will do! :)

[@mikecarr8199]

Hello, Thanks for your interesting videos. I hope I am not too off topic here: Do you know of any good conversational AIs that I can setup for some lonely elderly folk to interface with? The conversations need to be persistent. (the chat bot needs to remember names and what has been said...) It would be best if they were not connected to the Internet for privacy but if you have knowledge they are secure then they may be used. thanks for reading this and your time, Dziękuję Ci bardzo I do not speak much Polish but I am a fan of Poland. Please forgive me if you are not Polish.

[@karolinasowinska]

I'm super happy to hear that you're a fan of Poland! :) I'll try to look for something that will be relevant. There are plenty of chatbots available online, but I'll look for something with high accuracy. What's your coding experience btw?

[@mikecarr8199]

Thanks for your help.

[@karolinasowinska]

@@mikecarr8199 Sure, it's good to hear that you've good some experience. I'll have a look around tomorrow and let you know what I think good options are :)

[@karolinasowinska]

Hi Mike! I think that the best conversational chatbot at the moment is Replika. It's an app that you can download from an appstore. Perhaps try giving it a go and see if it does the job for you. But please bear in mind that deep conversations might still be a little off limits for the current state of technology.

[@mikecarr8199]

@@karolinasowinska Thank you for your help and advice. take care

[@jalajpratapsingh1625]

weirdo but real , appreciated !!

[@johnkenedy]

#ForTheEngagement

[@prod.kashkari3075]

Reinforcement learning is so hard

[@souptray1936]

So... did the guys give you the handout say No and leave? Sorry kind of want more of the story.

[@karolinasowinska]

Haha no, of course not. We first talked about my research. However, they knew more or less what it is about, because we exchanged a few messages before the meeting. So they came prepared. And they were nice to me (i.e. agreed to meet) because I was to be employed by that company, but in a different department.

[@farmdve]

Wow, I couldn't have imagined you'd be interested in studying the stock market. There was some research published recently that says markets are not efficient which means you can beat the market. I haven't had that success.

[@aleksandarrusinov9974]

When is model explainability complete? Can you transfer the model explainability problem to the halting problem - www.analyticsinsight.net/machine-learning-in-light-of-godels-incompleteness-theorems/

[@antoinefdu]

They told you that "you can't invest your money based on a black box"? That's surprising. Let's take an extreme example. Imagine that you let your algorithm run for a couple of months, and during these 2 months it has been able to systematically predict changes on the stock market, so much so that if someone started using it as soon as it was available, they would have made billions by now. How long do you think that "never use a black box algorithm" policy is gonna last?

[@karolinasowinska]

Your reasoning makes sense. In the real world full of regulations, watchdogs, etc., things can be hard to implement though. I don't know when (and if) this will change.

[@antoinefdu]

@@karolinasowinska You're right, I didn't think about that.

[@maximpobihun5469]

I still don't understand, why those guys from the big office wanted to meet with you?

[@karolinasowinska]

I was to be employed in another department of that firm, so they wanted to be nice for a future, junior colleague (I guess!)

[@yomismo5711]

... superbeautiful girl! 😍🌹

[@karolinasowinska]

How flattering! :)

[@khalidsultani6006]

she didn't know how she is beautiful

[@bermchasin]

@7:38 Im watching a year since this video has been released, but I am also an alien, so I wont comment.

[@jeffreylim5920]

XAI !

[@shawncook5083]

hahaha here in America people from my generation and down don't even know the concept of a pension anymore!

[@candyfloss184]

Hi Karolina....how are you?

[@karolinasowinska]

Hey, good thanks! How are you these days?

[@candyfloss184]

@@karolinasowinska I fell sick. Too many incidents raised by business users for our team. Too much work. Too much worries.

[@karolinasowinska]

@@candyfloss184 I'm sorry to hear that. I hope you'll get better soon!

[@candyfloss184]

@@karolinasowinska I feel nice to be able to communicate with you. I wish you have a wonderful day ahead.

[@bangnl]

It's hard 😂

[@easyphysi5568]

Can you become my guide...??

[@weoncualquiera8340]

Come to Chile! we could defend the falklands together jajaja

[@duckmeat4674]

🇬🇧♥️🇫🇰

[@karolinasowinska]

Hahaha. Isn't it too late for that?

[@soumiksaha]

soumiksaha.live Ur feedback pls

[@soumiksaha]

I love your videos but you are becoming pale and thin please take care of yourselves.

[@karolinasowinska]

I'm really alright, I think it's just the thumbnail picture that maybe makes me look pale! Thanks tho!