Analysis of ML and NLP publication statistics from 2019. marekrei.com/blog/ml-and-nl… #machinelearning #NLProc

While it's great to see new AI algorithms using human intelligence as the baseline, I just want to bring back this Edsger W. Dijkstra's quote: "The question of whether a computer can think is no more interesting than the question of whether a submarine can swim."

TechWolf vindt een miljoen voor systeem dat talent in kaart brengt op basis van AI Jeroen Van Hautte, C. Vandendriessche, Showpad #Startup #MondayMotivation #AI #HRtech hrmagazine.stry.link/m/AyO6qF

Congratulations to Guy Emerson on being Highly Commended for the CPHC/BCS Distinguished Dissertation Award! You can read an extended abstract of his dissertation here: cl.cam.ac.uk/~gete2/thesis_… bcs.org/more/about-us/…

Today would have been graduation day at St Catharine's College so to show their pride and appreciation to all their graduands they wrote it in the sky above their college. St Catharine's College #love #cambridgeclasof2020 #cambridgegraduation

Some musings on NLP/ML self-study. tldr: * Reading a paper is really hard, give yourself a lot of time and slack. * Pick depth over breadth. If you do depth well, you get breadth for free. * Be actively skeptical of all claims. Read your own drafts with 2x that skepticism.

International Baccalaureate program uses prediction algorithm to replace high-stakes exam. So much wrong with this; not just black box, but worse: deep misunderstandings about what you can expect and allow from ML by decision makers (incl at colleges). wired.com/story/algorith…

This is mind blowing. With GPT-3, I built a layout generator where you just describe any layout you want, and it generates the JSX code for you. W H A T

Still cropping and modifying BERT diagrams from Devlin et al. (2019)? I spent several hours redrawing BERT in PowerPoint so you don't have to... Perfect for use in presentations, papers, etc.! Happy to share under Releasing under CC BY 4.0 cs.uwaterloo.ca/~jimmylin/BERT…

GPT-3 writing React and SQL is the "neural style transfer" of 2020. Remember when these pictures were proof that AI understands art?

I have a joke about Gaussian processes, but you probably won't like it.

I'm not putting them on blast b/c they're a student, but I just ran across someone implying that GPT-3 correctly answered a medical question by using reasoning and underlying knowledge. Language models simply do not do that. PLEASE don't use them for medical advice.

For those of you who don't know Cambridge well - this is in the city centre.

Grammatical Error Correction in Low Error Density Domains: A New Benchmark and Analyses. Our #EMNLP2020 paper releasing a new type of error correction dataset, focusing on realistic web text. Join for the virtual poster session, 18 Nov 18:00 UTC. #NLProc

PhD student Josef Valvoda gives today's Tech Talk 'Deep Learning for Career Prediction'. A visiting researcher at TechWolf, which works on AI-based strategic workforce planning, he'll talk on its research into modelling the language of career sequences. ow.ly/pRbI50Du3U0

I’ve recently struggled to understand Python concurrency: - How threads & processes work on your OS vs Python - The role of CPUs, hardware & the GIL - Understanding beyond rules of thumb After tons of research, I wrote this w/the answers plus more 👇🧵 python.hamel.dev/concurrency/

You can now see a graphic representation of your .gitlab-ci.yml configuration. ➡️ bit.ly/2Oo9sA1

Very happy to receive an award for Young IT Scientist of the Year from the President of Estonia. :) Many thanks to all my colleagues at Imperial College London, University of Cambridge, Tallinn University of Technology and Transformative AI, along with all my great collaborators!

Very few people realize that there’s a massive tech paradigm switch happening between old-school ML (dare I say “big data”?) and new generation ML (transfer learning). All the tools/platforms/processes will be different from one to the other.