HOT TOPIC The AI Entrepreneurship thread. Code AI? Interested in AI? Data Science? Machine Learning?

[SamHalen]

So guys where do you see industry going in the upcoming years? Is it worth to dig deep into AI/Data Science if you will want to create something (considering you aren't a huge company like Google, Amazon etc.) and you're building from a ground up. Or is it better to go more into creating hardware/IOT stuff that will be needed for computation and also to gather data? If you would decide now to study one of them, which one you would go to?

The "frontiers" of AI are light years ahead of what you and I can access from online courses. Source: I manage an online course and it's taken me months just to create tutorials. I've heard of a few successful AI startups but they mostly get VC funding and have founders who went to Ivy league schools.

A more "fastlane" approach is to feed the crowd. What are the needs that all of these AI companies and people will need? All of these are unattractive, which means that there's less competition.

Skills

Sales and marketing of machine learning models

• How can you prove that your model is worth $100/customer? Data scientists tend to come from technical backgrounds. I worked as a predictive modeler who would create the models and PowerPoint decks, but then we'd turn them over to our sales team who would sell them to clients.
• My current course has about 30% of the curriculum just on communication. Basic things like writing, combining graphs/statistics into understandable language. The person who can explain their algorithm/AI wizardry in simpler language is worth more money than the person who struggles with communication.

Risks to fraud, data breaches, technical errors

• Data cleaning, integrity, and reducing error. How can you know that your data is going to be consistent? "Garbage In; Garbage out." Data scientists need to know this, but it's not new! It's been around in Database language since SQL was invented in the 1980s... but people don't want to learn it because it's "boring".
• Writing documentation.... Ugh. Is there anything that a data scientist hates more than doing this? Probably not. This task tends to get outsources to junior team members because it sucks. But the probme is that only the "expert" who created the model is knowleagable enough to write good documentation.

Hardware

• How to use cloud services. There are basically three big competitors - Microsoft Azure, Google Cloud, and Amazon Web Services.
  • Companies concerned about data privacy still are hesitent to use the cloud, unless they are massive and have boat loads of money to spend on R&D.
  • Getting models into "Production". It's easy for someone to take their Python or R and make some AI tool, but it takes a lot longer to get it ready to be used in real-time. Most of the time this needs to be translated from R/Python into some other framework like JavaScript/Scale (?), Julia, etc. This is one of those tasks that data scientists hate.
  • Budgeting... The cloud offers "unlimited scalability" but with that comes unlimited costs. At my former company, we ran into major problems with cost overruns. We were using a machine learning model which was designed to run on about 100 GB of data, but then we had a client that had 100TB of data. The cloud computing costs went from about 20% of the costs (for normal clients) to 60% of the costs. This could have been avoided if we had knowledge of how to manage costs.

Hiring/Staffing/Recruiting

• How do we hire the person with the right credentials?
  • Hire only Phd's
  • Hire Master's degree people and hope that their degree is actually useful
  • Put them through a rigourous hiring process involving technical projects, background checks with previous managers, reviews of their github/kaggle projects, and numerous on-site interviews and "white board" questions
• How do we know if a "data scientist" from India who is selling themselves on Upwork for $500/hr is legit or not?

Education

• Analytics managers have "continuing education" requirements like doctor's do. As algorithms get more complicated, it becomes more difficult to keep staff up to date. There's a growing need for corporate training. Remember all of that "Data Privacy" training that you were required to take? Imagine "AI Privacy" training in 3 years.
• Learning platforms - My site is build using Moodle, which is good because it has a ton of open source documentation.. but there are newer platforms which have more features such as zoom integration, templated content (think Wordpress) for quickly making new courses, and so forth.

 

Don't like ads? Remove them while supporting the forum. Subscribe.

[ygtrhos]

I did not read the whole thread and I have a question.

I am a 31 year old simulation engineer. Basically I do technical simulations.

For example, if you have a connector for a pump and want to calculate the loads on the bolts at the connector, I can calculate it.

Good thing is, that I do not actually need to click around. I can sit down and program something, which actually builds that model and creates the code. It has been a while, but I can program in Pyhton, MATLAB and C.

Now recently, I have seen a video that shows the capabilities of this new GPT-3 code. It has read all Wikipedia and books that have been published since Earth existed or whatever and now it can compose its own poems.

Heck, it also writes the code if you say "write me a code that rolls a dice with 20 sides and records the first 10 throws". Or "make a mockup of a website that is a photo gallery where you can like or comment to each photo" (Instagram) and it makes a mockup of Instagram in minutes.

This brought me to the idea that I can apply this in my own field. I cost 100k€ to my company and there are probably like 100k-300k guys like me.

If I really can integrate AI to my own field, I would be spectacularly rich. You would not need then hours of work and 20 engineers in an office, but 2-3 very qualified ones would be enough.

Practically this would be like "calculate connector loads with 6 X bolts under Y kN of bolt pretension and the connector's drawing is in Z file."

This is an objective that requires creativity, knowledge and experience, but it is not dependent on emotions or emphaty, so I think it is possible.

But I have zero experience with machine learning or AI. So I have some questions:

1. Where to start with AI where you can automatize such objectives? What should I learn or how should I qualify myself?

2. I see a big challenge: Whereas you can find a lot of HTML code or poems on the internet, there is limited work on coding such specialized engineering software because most of such codes rot away in archives of geeks like me. The community is very small and very much disconnected, so such codes to sample from are not easy to acquire, I believe.

I heard that the bigger the data size is, the better. Is the "artificial intelligence" and the data size linearly proportionate? Or exponentially?

I hope I can convey where I am going with those thoughts. My concern is that it might not be easy to feed the code enough data.

I would appreciate any answer.

 

[sonny_1080]

Attorney General Anne Milgram uses "Moneyball"-esque techniques to reform the criminal justice system, with great success.

Anne Milgram: Why smart statistics are the key to fighting crime

When she became the attorney general of New Jersey in 2007, Anne Milgram quickly discovered a few startling facts: not only did her team not really know who they were putting in jail, but they had no way of understanding if their decisions were actually making the public safer. And so began her ongoing, inspirational quest to bring data analytics and statistical analysis to the US criminal justice system.

That's the most inspiring thing I've heard today.

 

[ChrisV]

Great interactive Demo of how a Convolutional Neural Network works

CNN Explainer

An interactive visualization system designed to help non-experts learn about Convolutional Neural Networks (CNNs).

poloclub.github.io

 

[ChrisV]

Great interactive Demo of how a Convolutional Neural Network works

CNN Explainer

An interactive visualization system designed to help non-experts learn about Convolutional Neural Networks (CNNs).

poloclub.github.io

Two more interactive Demos on Decision Trees:

A visual introduction to machine learning

What is machine learning? See how it works with our animated data visualization.

www.r2d3.us

A visual introduction to machine learning, Part II

Learn about bias and variance in our second animated data visualization.

www.r2d3.us

 

[Martin.G]

Nice resource you share @ChrisV.

I am reading Reinforcement learning: an introduction (Richard S. Sutton, Andrew G. Barto), a great book that cover all what you want to know about the subject, from the beginning of this discipline to today.

 

[lowtek]

Nice resource you share @ChrisV.

I am reading Reinforcement learning: an introduction (Richard S. Sutton, Andrew G. Barto), a great book that cover all what you want to know about the subject, from the beginning of this discipline to today.

This is a fantastic text. I refer to it frequently.

 

[ChrisV]

South Park creators Trey Parker and Matt Stone created a new AI-based television show "Sassy Justice," and it's great.

 

[srodrigo]

I ended up doing it. I'm half way through. It's a good introduction, although it looks like I'd need a real project to work on later.

Just about to finish the Deep Learning certification on Coursera (managed to go back into it a year later and finish courses 4 and 5). I have mixed feelings about this:

• The content is good, but the instructor sounds like my maths teacher (yeah, I get that this topic is about maths, but you know what I mean). It's difficult to keep focused without falling asleep.
• Video edition is non existing. The guy speaks so low at the end of the sentences, that it's difficult to understand what he's saying. It doesn't help that he doesn't use a proper microphone either. Not to mention that mistakes and repetitions are not even removed. A joke.
• I got sick of cats and cars images (one of the reasons why I got bored and did quit last year). There is a bit more variety towards the end though.
• These courses are a good introduction, but I don't think you can jump into a job after them. I feel there's a lack of more challenging, real-world projects such as (from what I've heard) the ones on Udacity nanodegrees.

Happy to be nearly done and finish something, but I feel like I didn't learn much from it.

 

[ChrisV]

Just about to finish the Deep Learning certification on Coursera (managed to go back into it a year later and finish courses 4 and 5). I have mixed feelings about this:

• The content is good, but the instructor sounds like my maths teacher (yeah, I get that this topic is about maths, but you know what I mean). It's difficult to keep focused without falling asleep.
• Video edition is non existing. The guy speaks so low at the end of the sentences, that it's difficult to understand what he's saying. It doesn't help that he doesn't use a proper microphone either. Not to mention that mistakes and repetitions are not even removed. A joke.
• I got sick of cats and cars images (one of the reasons why I got bored and did quit last year). There is a bit more variety towards the end though.
• These courses are a good introduction, but I don't think you can jump into a job after them. I feel there's a lack of more challenging, real-world projects such as (from what I've heard) the ones on Udacity nanodegrees.

Happy to be nearly done and finish something, but I feel like I didn't learn much from it.

Honestly I feel that way about most courses. The Coursera one and fast.ai were ZzZzZzzzz

The only courses that worked for me were Kirill Eremenko's. His courses are on Udemy snd SuperDataScience.com

Kirill Eremenko | Data Scientist| Udemy

Kirill Eremenko is a Udemy instructor with educational courses available for enrollment. Check out the latest courses taught by Kirill Eremenko

www.udemy.com

SuperDataScience

www.superdatascience.com

For R the tidyverse / tidymodels websites are great:

Get Started - Build a model

Get started by learning how to specify and train a model using tidymodels.

www.tidymodels.org

In general it's better to read official documentation like that than reading dumb tutorials people put up.

Oh and another essential tip: find a community where you can ask questions. Discord is probably the best, but there are also AI Facebook groups, AI/ML subreddits... just somewhere you can ask questions if you get stuck.

 

[Martin.G]

Honestly I feel that way about most courses. The Coursera one and fast.ai were ZzZzZzzzz

The only courses that worked for me were Kirill Eremenko's. His courses are on Udemy snd SuperDataScience.com

Kirill Eremenko | Data Scientist| Udemy

Kirill Eremenko is a Udemy instructor with educational courses available for enrollment. Check out the latest courses taught by Kirill Eremenko

www.udemy.com

SuperDataScience

www.superdatascience.com

For R the tidyverse / tidymodels websites are great:

Get Started - Build a model

Get started by learning how to specify and train a model using tidymodels.

www.tidymodels.org

There are very good courses here. This is gold.

 

Don't like ads? Remove them while supporting the forum. Subscribe.

[ChrisV]

I just remembered I wrote this article a while ago:

The absolute fastest way to learn R for Data Science.

This post will be updated as times goes on. But for now, the most important things are:

medium.com

Data Science and Artificial Intelligence are distinct but similar fields, so the information in that article can also help you with AI / Machine Learning.





It's based on the R programming language, but many of the suggestions can be applied to python as well.

The best way to learn something is to learn the top ~20% of most important things, which will generally get you 80% of the results. In the article I suggest you just do it. Don't waste time learning a shitload of theory at first. You can learn that later. It's best to focus on the actual things that professionals use.



Jeremy Howard likens learning AI theory to learning baseball. Imagine if you wanted to learn to play baseball. But instead of just going out and playing baseball they sat you in a classroom and taught you the physics of the game. They taught you equations and theory and made you memorize it. This is obviously a really silly way to learn and by the time you got to even play a game you would have probably lost interest. He talks about it here.

How this relates to my article is: the best way to learn Data Science or Machine Learning is to do Data Science or Machine Learning. Kaggle.com has all sorts of datasets you can download and experiment on. That way by actually doing the work you learn the most important things used in production (the 20%.)

Some notable tips from my article:

Write the code. Don't copy and paste code you find in tutorials or StackOverflow. Make sure you actually physically type the code. You need the muscle memory. and you need feedback on what you're doing right or wrong. When programming it's so easy to put a period or closing bracket in the wrong place and unless you're actually typing the code you'll never know what you're doing wrong. Act like Copy and Pasting is a felony.

When you learn something: put it into practice immediately. Our cognitive memory is terrible, but out experiential memory is great.

Things need to be ingrained in your mind. You create different neural pathways when you do something. You don’t even necessarily have to intellectually know something to have the neural pathways. Have you ever needed to give someone a phone number, but without the number pad in front of you, you couldn’t remember it? When there’s a phone in front of you. you can dial the phone number just fine, but when someone asks you you’re like ‘uhhhh, let me go look at a number pad.’ Or the same with directions sometimes. You cant give someone directions but you ‘know the way if I just drove there.’ This is why doing is so important.

You have a muscle memory where you know exactly what you’re doing…. but you can’t intellectualize it. And the opposite can happen; you can intellectualize something, but have no idea how to actually do it. Don’t make that mistake!

You need to run the command immediately after entering to see if there are any errors. You absolutely need feedback. If you type it into the script, and don’t enter the command, you can easily engrain an incorrect syntax into your brain over and over.

Here are some evidence-based strategies for the best ways to learn things:



Btw, I have a bunch of life-improvement related articles on Medium as well.

Chris Vaccaro – Medium

Read writing from Chris Vaccaro on Medium. Evidence-based insights. Every day, Chris Vaccaro and thousands of other voices read, write, and share important stories on Medium.

chrisvacc.medium.com