As a Data Scientist or a Machine Learning Engineer which language do you prefer?

Is it R or Python or any other language? I prefer Python. I use Spyder as an IDE for my projects. What are your thoughts on Deep Learning, Neural Networks and AI?

This page is a permanent link to the reply below and its nested replies. See all post replies »

iMystery · M

I find AI fascinating as hell!... but the risks of malfunctions, hacking and programme corruption (through hacking) cannot be ignored.

SW-User

@iMystery can you hack the "black box" ?

Beardoo · 31-35, M

@iMystery Risks of what? I don't think hacking is related here bro

iMystery · M

@SW-User I am no hacker lol. So wont know where to start... but I believe people would go to great lengths to find some way of hacking AI.

iMystery · M

@Beardoo So software used in AI cannot be somehow corrupted?

Let me mention that I am no programmer or data scientist... so my question is sincere.

Beardoo · 31-35, M

AI is actually to help humans. See, AI would just ease out and automate our work. It won't destroy the world as they show in movies.

SW-User

@iMystery me neither! Just asking.

SW-User

@Beardoo what if someone changed the real world model of some general AI. Certainly some believe access to the internet would have a negative out come to general AI?

Beardoo · 31-35, M

@iMystery First of all it's not a software bro. It's an algorithm. Secondly, it depends on the coder what he is using it for. So by that means, it can be used for multitude of purposes which can be not be always right in the moral sense too. However, it's not related to hacking.

iMystery · M

@Beardoo

AI would just ease out and automate our work

This too would be a concern of many... people losing their jobs to AI.

Beardoo · 31-35, M

@SW-User I agree with you. We as of now, don't know to what extent it could go. And yeah, model with adapt to the real world data.

Beardoo · 31-35, M

@iMystery For this bro, new jobs would be created. As of now, AI is just a model. Like for example, Doctors using AI for cancer prediction. That doesn't mean doctors will go in vain. You need a human to put those parameters and also the data. For example, an AI might find relationship between cancer and your salary if salary of the people are mentioned in the dataset and also the salaries are coincidentally same for people having cancer. That in the right sense is wrong, but AI doesn't understand that. It only tries to find pattern in the data the doctor provides. So it's the job of the doctor to remove all the irrelevant parameters so that the predictions our model make are efficient.

So we would definitely need humans and new kind of jobs would be created for the same.

Beardoo · 31-35, M

@iMystery I would suggest trying the Andrew Ng course of Machine Learning with Stanford University on Coursera. It's an awesome course and you will get to know how things actually work.

iMystery · M

@Beardoo I will look into that some time.