Andrew Ng: Advice on Getting Started in Deep Learning | AI Podcast Clips
Вставка
- Опубліковано 20 лют 2020
- Full episode with Andrew Ng (Feb 2020): • Andrew Ng: Deep Learni...
Clips channel (Lex Clips): / lexclips
Main channel (Lex Fridman): / lexfridman
(more links below)
Podcast full episodes playlist:
• Lex Fridman Podcast
Podcasts clips playlist:
• Lex Fridman Podcast Clips
Podcast website:
lexfridman.com/ai
Podcast on Apple Podcasts (iTunes):
apple.co/2lwqZIr
Podcast on Spotify:
spoti.fi/2nEwCF8
Podcast RSS:
lexfridman.com/category/ai/feed/
Andrew Ng is one of the most impactful educators, researchers, innovators, and leaders in artificial intelligence and technology space in general. He co-founded Coursera and Google Brain, launched deeplearning.ai, Landing.ai, and the AI fund, and was the Chief Scientist at Baidu. As a Stanford professor, and with Coursera and deeplearning.ai, he has helped educate and inspire millions of students including me.
Subscribe to this UA-cam channel or connect on:
- Twitter: / lexfridman
- LinkedIn: / lexfridman
- Facebook: / lexfridman
- Instagram: / lexfridman
- Medium: / lexfridman
- Support on Patreon: / lexfridman - Наука та технологія
the aura of andrew ng is so peaceful. really inspiring podcast.
He is Singaporean... its kind of representative.
i dont mean to be off topic but does someone know of a tool to get back into an instagram account??
I was stupid forgot the login password. I love any tips you can offer me!
His voice is smooth like satin. This is the first time I am hearing to his voice…😅
Can you Imagine a boy from very rural India where no electricity. No concret road. He is very much interested on AI now listning your every Interview very mindfully.
same here
👍
Me too
If he doesn't have electricity, how is he supposed to charge the device needed to watch this?
Just saying 🤔
@@dec13666 may be he is deprived of continuous power supply. Because it is major problem in rural area. Some villages get about 8 to 12 hours power supply in a day.
OMG hearing his voice again is like going back to those days when I was watching his tutorials in order to start with ML. Andrew helped a lot of people and most of the time, when I write something related to ML, I refer to him as the reason of why I am able to explain some of my knowledge. An absolute legend.
BTW he is totally right about taking notes in a paper in order to retain information. That’s what is helping me more in understanding the logic behind algorithms and coding challenges asked by big tech companies, because I have to admit that I am very bad at those 😆
I'll start taking hand written notes right now! I literally SUCK at algorithms and data structures
At what minute of this video he talks about taking notes? Thank you kindly.
are u using linkedin i'd like to connect
@@raznatovicanastasija 16:19
Feels so good to see Lex surprised and happy in a sense after Andrew asked him something
Andrews ng is the one of the greatest teacher in this era !!!
I've no words to express my gratitude for such humble intelligent generous person. Thanks prof Andrew Ng.
Andrew Ng is a perfect teacher. The way he teaches is great.
When the interviewee and the interviewer master the subject... Great job Lex and always exciting to come across any content/material from you. Thank you gents
Stephen is an amazing teacher. "Every word needs to fight for it's life".
man, this interview is pure gold, thanks for sharing
met him today in Portland Oregon.
I've been taking his courses for the last 8 months. I've been already following his career. He is just as humble and cool in real life. He's the GOAT
Andrew ng is the person who made me fall in love with the subject. Me from a biology background started his coursera just for a small project with no background in maths and code. The way he presented the mathematical basis was so motivating that I was building concepts from the framework he built now I am more interested in learning the math behind it.
Thanks for arranging this. Much appreciation 💝
Andrew Ng is inspiring, peaceful, and incredibly polite.
This is my second all-nighter in a row binge watching Lex Fridman Podcast videos, and I can say that I need another brain to help me comprehend all the insights I have gained.
Thanks Andrew Ng and Lex. Both of you are doing great work.
I see so much well-deserved love and respect for Andrew Ng in these comments, but props to Lex also for asking all the right questions. Amazing interview!
This guy Andrew has a sense of calmness and he is so humble. I did his beginner level AI 4 Everyone course and he is such a nice guy to hear to.
Thanks Lex for all the great videos. I'm not really a computer science guy and am wondering if you'll ever have more videos on applied mathematics. I especially liked the video with Professor Strang. I used his textbook in my Linear Algebra course many years ago. Thanks and all the best.
Thanks for sharing such interviews!! (:
yeeeeeah big thank you to this man. i love this man he helped tones of people on there project with his videos
Thank you for this video much appreciated bro
Andrew Ng's course on Machine Learning is AMAZING. I took it 8 years ago, and took it again ~3 years ago.
Thanks Lex and Andrew for sharing this conversation. Great insights !!
Excellent interview thank you!
Wow, this gentleman is incredible. I love that he takes human learning as seriously as ML.
Lex Fridman, this channel is a gold mine. Thank you.
fr
Andrew is a genuinely good guy and an excellent teacher.
Thanks for listening to your audience and creating these clips. Direct, informative - perfect.
Thank You dude! Keep rocking
lmao though 26 mins is still way to long, but I suppose for the content it's right.
BTW who switches the camera? Is it AI or a clone Jamie?
I loved that you mentioned knowledge graphs. I've been looking into them and I find them neat.
Prof. Andrew Ng has helped millions of students around the world by making his tutorials available to everyone on internet. Thank you!
PS: He is so right about getting stuck at a "reputed" company/university and where you don't enjoy what you're doing.
Andrew really cares a great human waheguru bless him love the man
Thank you! Alex, you are doing amazing work. Taking the course ASAP.
who is Alex though..?
Your channel is a goldmine.
Keep it up
thank u for the great talk !
Finally someone who says you dont need a phd in mathematics statistics electrical engineering and fucking physics. Ive been doing deep learning and data analytics for a bit with basic python amd mathematics and i agree fully with what Andrew started with
Amazing talk especially the latter half 👌👌 Thank you Andrew and Lex.
I always fascinated by this man's humility
i love andrew even hearing his voice takes me back to his tutorials.
Just signed up for the course. Excited to get started, thanks for sharing your knowledge!
Which programming language is this course based on ?
@@elpolodiablo5486 python
Andrew Ng and Lex Fridman. Two of main favourite AI minds! Legends!
Josh Starmer
I'm one of those guys who have benefitted from the Financial Aid option of Coursera and Deep Learning Specialization... And I'm very indebted to Andrew Ng for that.. Hope I'll teach more unknown people about it and give back as much as I can
yes to that mindset !
Bro how was that certificate useful, excluding the knowledge you gained... Do companies here in India value those certificates from Coursera, edx... etc?
@@Abhi-ct1gp no they don't value any certificates/specializations. But they do value your knowledge from these certificates and projects
Hi Siddharth, I have a few comments that may or not be useful to you and others. In case you are ever worried about financial aid in those Coursera courses know this : completion of the course (officially) whereby you receive a certification for completion is absolutely meaningless when finding work. I have interviewed dozens of students for AI related work, and in my opinion, the industry is flooded with these certifications from all sorts of online courses. On coursera you have the option to audit the course for free, which gives you access to all the material just not the certificate. I have seen so many students with certificates that have no real world experience, but the best students are always those that took what they learn and applied it to a small project of their own - however small. Never feel like the financial obligation should hold you back, and always know its not the certificate, its what you can do with the knowledge that matters. Hope that helps.
@@kenbobcorn You are absolutely right about the certificate part.. The assignments are actually very dumbed down imo.. Whenever I worked on a project or internship.. The most difficult part was creating the dataset in a format which can be fed into the model.. And that part is already done in all the assignments
His course on ML is one of the best classes I ever took. Highly recommend.
Awesome interview love it.!
Very good video, thank you so much!
Great. Thank you for doing this.
Andrew Ng seems great educator tackled even way of learning by make it a habit from few technics like writing notes. It is true to me, writing notes those difficult parts from learning will help to hard embed those informations in your head for easy recall. What he said is something quantum mechanics which where you put your attention is where your energy is.
Such a great inspirational video, Andrew Ng is so cool...
Thank you, Lex.
Great session
Hey this dude came on the Podcast to remind to continue the Coursera AI Course by Standford haha
Sir, You don't ever need to doubt yourself on whether your course/specialization is the best or not. they absolutely are the best.
That watch. I love that he shows it off :)
Wow! This is the best advice I've ever heard! Thanks Andrew Ng and Lex!
Chrys R what was that?
@@rezajanjani7712 I'm glad you asked!
Here's everything I found extremely useful:
1) Make learning( or studying, I prefer learning since studying is something I attribute to school whereas learning is for personal development) a habit.
2) Sustained effort over a period of time is way better than a burst of effort, in the long run
3)While taking a course online, take notes to retain the information.
4) With regards to a career in Machine Learning :
a) Get your foundation strong i.e be thorough with the basics and only then start working on the big things
b) find a passion project, no matter how simple to apply what you learnt
c)Dont go to a place( either to work or study) just because of the name on the building but instead see if the people and the culture align with your principles.
Andrew Ng is the god of this subject.Period.
@Lex, Your videos are inspiring. Your thought process about the world and philosophical mind is mind blowing 😍 !
Thank you !
Nice to hear your voice Andrew from high quality microphone finally 😁
I have learnt more in these 26 minutes than I learnt in my whole previous semester.
that's true
LOVE THIS!
I am taking this specialization on Coursera, this interview was so helpful.
Andrew gives brilliant advice
I agree to the RL part completely!!
Reinforcment learning is a part that really inspired me,
The Utopia of ml the future of ml
Thanks to Lex and Andrew for such a valuable interview!
Andrew ng is the best teacher
I took Andrew's course on ML using Matlab 2 years ago, I hoped for something from him related to Python, and he made full specialization for it. Definitely taking the course as soon as I get some free time.
Andrew Ng course made me fall in love with AI especially his machine learning course
the moment was when in the 5th week while showing the potential of neural network there comes a slide showing a man trying to gain vision using a camera and transmitting the electrical signal through his tongue. My friend that was the moment I knew that's it. that's what I will be working on.
[17:10] «Recoding knowledge in your own words promotes long-term retention. Comparing handwritten notes and typing, you can usually type faster than you can handwrite notes, so when people type, they are more likely to just transcribe verbatim what they've heard and that reduces the amount of recoding.»
I feel like that's because typing, especially if you're a really fast typer, is more of an unconscious action. Whereas writing is a more deliberate action.
Regarding handwritten notes: I've found that it is even better if you illustrate the concepts you are learning.
Thanks andrew
I utterly agree on what Lex mentioned, considering the Reinforcement Learning and how it hasn't been fully explored, hence there are no notable applications using it. The biggest application i can think of, in terms of utilizing Reinforcement Learning, is OpenAI's Five. I am highly looking forward seeing more real world applications.
And two years later. ChatGPT.
Thanks, Lex for the great effort you're putting on this program. I have to say that I expected the interview with Andrew Ng to be more interesting, you're interviewing one of the most interesting people and the leaders of the AI domain, yet you spent the whole interview asking educational questions about the courses and the learning prosses and pre-requisites instead of asking the meaningful, important questions you usually ask other guests! I hope you take this as positive feedback than criticism.
Andrew Ng i am indebted to you.
When Andrew Ng started talking, I went into the AI learning mode straight away.
The greatest minds on 🌎 We Love you ❤️
Отличное интервью
So many life lessons here.
Andrew Ng is so cool. ♥
I love it
Andrew is best
Two kings
andrew ng is so zen
I see Andrew I click
finished watching
Lex: *asks question*
Andrew Ng: *replies with same question*
Lex: "I think that's a really good question"
Teaching is an art. Gratitude to Andrew neg
Sorry in advance for asking such a basic question.
Can you suggest me places where I can find good research papers on AI / ML or Deep Learning? Thank you.
Yes, take a note in the paper when attending the course in Online, They Can Very Well to Increased your Memory Because When you Take a not in Paper Your Mind Focus on the particular point who you write on paper and never you forget who write on paper, it my Personal Point to Share, I am a Btech Student of Machine Learning Engineer.
Andrew Ng lectures help me so much with my Dissertation.
This felt like a police custody interview in a room where Andrew Ng was the criminal and this guy was the chief incharge of him
I've been doing his course as well . Hehe . I'm so used to his voice .
I could listen to that guy all day! Would have loved to hear KAGGLE once!
Is there any comprehensive math course for ML that covers the topics in their entirety (no pre-requisites needed)?
Wise men.
He says dater sometimes, in ML course I thought it was the recording device but he does say it. Safe to say ;
from AndrewNG import data as dater
from AndrewNG import StanfordOnline as so
for data in so :
if (too much dater)
collect(dater) = False
else:
collect(dater) = True
This guys machine learning course was the way I started also.
great
If you're just getting started in the field, it's also worth considering exploring AI outside of deep learning! These days, machine learning is hyper-competitive, results change fast, and it can be hard to catch up to the "state-of-the-art", especially if you don't have a job at a company with massive troves of data.
Right now we're in an age where deep learning is king, and many are predicting it will be the only form of artificial intelligence in the future. If you study the history of AI you realize hype comes and goes, but the technologies underneath survive and evolve over time. Symbolic methods, often referred to as "Good Old-Fashioned AI" or GOFAI never died, it has evolved into a rich intersection of neuro-symbolic reasoning, search algorithms, task and motion planning, robotics, etc.
These approaches are not entirely separate from deep learning, and generative modeling is an emerging unifying paradigm for both symbolic and neural approaches. Many will be coming from the neural-network side and trying to figure out symbolic methods. In that space, it will be extremely hard to get a competitive edge. Correspondingly, there is an extremely well-placed niche for people working on symbolic methods (especially those who are well-versed in probability) to make a huge impact on neural networks. If you've read this far, I'd especially recommend looking at probabilistic programming (disclosure, that's what I work on) to expand your perspective of what sorts of AI are possible.
I have a job in Deep Learning thanks to Andrew Ng