Psychology, acoustics and mathematics - AI and big data

Anything to do with the traditional world of get a degree, get a job as well as its alternatives
Post Reply
guitarplayer
Posts: 1303
Joined: Thu Feb 27, 2020 6:43 pm
Location: Scotland

Psychology, acoustics and mathematics - AI and big data

Post by guitarplayer »

I am doing a BSc in Mathematics and Statistics at the moment. In the first assignment I have been asked to write a bit about myself. I wrote that I did one year of uni level physics (acoustics), an MA psychology with quite a bit of statistics, and worked as a market research analyst but had been in the care sector for the past 5 years.

My tutor said that if I have any interest in AI and / or big data, these are growing fields and with lucrative and interesting jobs, and it could be relatively easy for me to get in.

Anyone here works in AI and could advice if I should look in this direction? UK/EU centred advices particularly (though not exclusively) welcome.

User avatar
Lemur
Posts: 1613
Joined: Sun Jun 12, 2016 1:40 am
Location: USA

Re: Psychology, acoustics and mathematics - AI and big data

Post by Lemur »

I don't work in AI, but Big Data is definitely a growing field.
Also cloud computing, AI/Machine Learning are growing and really heavy on mathematics. AWS Cloud / MSFT Azure.

Quadalupe
Posts: 268
Joined: Fri Jan 23, 2015 4:56 am
Location: the Netherlands

Re: Psychology, acoustics and mathematics - AI and big data

Post by Quadalupe »

Most companies that want to do 'something' with AI or data science are mostly talking about making dashboards or some very simple predictive models. You'll be hailed as the excel macro messiah, but you will probably be bored to tears. So be aware of where you apply! If you do some online courses on coursera, udemy or fast.ai and complete some small projects, you'll be good enough (especially with a math background) for 90% of the jobs.

Also, if you gain some experience in 'wrapping' a trained model in a simple Flask application, so that you can actually deploy it to a server, you'll be beating a least 50% of your competitors for the job. Most of the courses only focus on training a model and optimizing the metrics, and not on what you'll actually do with the model once you have it.

If you have any further questions, feel free to ask!

guitarplayer
Posts: 1303
Joined: Thu Feb 27, 2020 6:43 pm
Location: Scotland

Re: Psychology, acoustics and mathematics - AI and big data

Post by guitarplayer »

@Lemur, @Quadalupe thanks for your thoughts.

One step I am going to make in this direction will be to change one elective I get to have early next year from 'Economics in Context' which is history related and UK based to 'Technologies in Practice', which is about CISCO, Linux and AI / Robotics.

Quadalupe
Posts: 268
Joined: Fri Jan 23, 2015 4:56 am
Location: the Netherlands

Re: Psychology, acoustics and mathematics - AI and big data

Post by Quadalupe »

I'd also encourage you to already look into some challenges on Kaggle. Try solving some by your own and then look at the winning submissions. You'll learn a ton! Also, you can quickly build a portfolio like this. Especially if you go the extra mile of deploying your solution somewhere where people can use your solution for inference. But don't focus on this for your first project, just give it a go and have fun!

bostonimproper
Posts: 581
Joined: Sun Jul 01, 2018 11:45 am

Re: Psychology, acoustics and mathematics - AI and big data

Post by bostonimproper »

If you are capable of programming and have a stats background, getting into machine learning is fairly straightforward. Check out the Google Machine Learning Crash Course.

TBH, most "AI and big data" is really spending 20% of time figure out what the problem the business side wants you to solve actually is, 65% of your time cleaning data sets, 10% of time hydrating a bunch of random low level features, and then 5% of your time sticking it into tensorflow and finding out, in fact, the best model continues to be a one-layer neural network (i.e. logistic regression).

From the people I know in London, UK salaries are low compared to the US. Most of my American colleagues are making 2-3x my UK colleagues at the same level of skill and experience. From what I understand Switzerland has much better pay?

guitarplayer
Posts: 1303
Joined: Thu Feb 27, 2020 6:43 pm
Location: Scotland

Re: Psychology, acoustics and mathematics - AI and big data

Post by guitarplayer »

@Quadalupe I did not know about Kaggle but DW knows about it. Thanks for the advice, I will have a look.

@bostonimproper thanks, I am going to check out this course. I am more or less what you have described, capable of programming (i.e. not doing it now) and have (some) stats background (as in stats from studying social sciences). But with time I might learn some programming and will surely get more stats background with what I am currently studying.

UK salaries are notoriously lower than in the US which is sad. I am reminded of it every time I read journals here from US based folk. This is also part of the reason why I still work in care which is considered low skilled (though requires lots of skill if you want to do it well) and low paid. The salary bump when moving to a white collar job would not be really that significant from my point of view, ditto savings (especially given my special living arrangement).

I have had Switzerland on my radar for a while now. I would not be able to work in care there because of the language barrier. I am pretty sure I could spot something analytical there though, especially with this BSc in Maths and Stats and maybe some experience in the field from the UK. It could be for a salary bump, but gosh, I think mostly for the Alps and better weather.

In general I am quite excited by the opportunities that this endeavour might bring me. https://www.smbc-comics.com/index.php?db=comics&id=2722 feels like starting another lifetime!

guitarplayer
Posts: 1303
Joined: Thu Feb 27, 2020 6:43 pm
Location: Scotland

Re: Psychology, acoustics and mathematics - AI and big data

Post by guitarplayer »

Hi all, fast forward three years I got my maths and stats degree and am working as a statistician (not bio / pharmaceutical). I know I will soon have an opportunity to get promoted and remain as a statistician, but I also see openings for operational researchers.

I completely see the merit of operational research on the private side of things, or for a working man, in the salary man setting I wonder if this the organisation's 'bad guy' (since to do with optimisation).

As a statistician, I am in a position converging to that described by @Quadalupe up in the third post. I am yet to get to the place where I am bored to tears, but I already get lots of kudos for answering simple questions.

I give myself a few more years doing what I am doing and think it would be a nice touch to maybe finish somewhere in the Alps so this also feeds into the equation - getting a job in Switzerland or around there from the position of a statistician or operational researcher, or it does not really matter.

I will talk to my manager about that opening soon. From the forum, any opinions about working in operational research?

Post Reply