it’s Sunday, it’s really windy weather so stay at home and rejoice because the new DSB arrived! My recommendation for this volume is an article with tips for programmers in Computer Science & Science or an interview with Mark Zuckerberg about metaverse in Business and Career. But there are so many other interesting articles like the one about the new concept of cookies by Google or interesting Reddit threads and more and more and many more!
As always, enjoy your reading.
https://lukewood.xyz/blog/keras-model-walkthrough – If you are into Keras, then have a look at this article about Model class and some of its mysteries. (rcmd by reader)
https://ruder.io/ml-highlights-2021 – Another summarization article, this time it’s about ML research highlights of 2021 and it’s worth your time.
https://towardsdatascience.com/asset2vec-turning-3d-objects-into-vectors-and-back-8335496b756d – Asset2Vec to create vector representation of 3D objects with NeRF. Simple intro to NeRF is here.
Computer Science & Science
https://matt-rickard.com/reflections-on-10-000-hours-of-programming/ – 31 tips for advanced programmers on how to be even better written by a guy who spent more than 10 000 hours coding and was a software engineer at Google. (rcmd by reader)
https://matt-rickard.com/runescape-machine-learning/ – Well, I have never knownRuneScape, but I found this brief article about object detection in the game quite catchy. Code is available here. (rcmd by reader)
https://www.ryanpickren.com/safari-uxss – How to hack the Apple Webcam. Author of this article was rewarded $100,500 as a bounty.
Business and Career
https://www.theverge.com/22588022/mark-zuckerberg-facebook-ceo-metaverse-interview – Interview with Mark Zuckerberg about the metaverse and future of the internet. Can be also listened to as a podcast.
https://www.bloomberg.com/news/articles/2022-01-28/orcel-says-unicredit-withdrawing-from-otkritie-due-diligence – The tension on Ukraine is preventing UniCredit from buying Otkritie Bank.
https://www.nytimes.com/2022/01/03/business/wall-street-cloud-computing.html – What is the potential of cloud technologies in the banking industry and why are the banks so slow in its adoption?
https://blog.seznam.cz/2022/01/neni-dotaz-jako-dotaz-aneb-jak-funguje-entitni-vyhledavani-na-seznamu/ – Jak udělat vyhledávání relevantním. (rcmd by reader)
https://medium.com/@david_colombo/how-i-got-access-to-25-teslas-around-the-world-by-accident-and-curiosity-8b9ef040a028 – 19 year old German was able to control more than 25 Tesla’s due to several security flaws in 3rd party software.
https://www.theverge.com/2022/1/25/22900567/google-floc-abandon-topics-api-cookies-tracking – Google is planning to replace cookies in Chrome with Topics API. And of course there are a lot of unhappy people because of that decision as is described in this article by Wired.
https://realpython.com/python-modulo-string-formatting/ – Modulo string formating in Python. Don’t underestimate it, it can do a lot of magic.
https://nickch-k.github.io/SomeThoughts/posts/2022-01-23-overdebunked/ – Be skeptical about statistics, but not too much! 6 examples where you should not be so harsh, like misunderstanding control variables or that correlation is not causation.
Data & Libraries
https://towardsdatascience.com/good-data-citizenship-doesnt-work-265f13a37fa5 – To properly manage data is a difficult task. Read why the documentation is not enough and much more important is organization (data owners, acquainted consumers etc).
https://marketplace.visualstudio.com/items?itemName=pcsoftware.readable – Readable = Ai si VisualStudio extension that automatically generates comments in code. (rcmd by reader)
https://www.montecarlodata.com/blog-the-non-engineers-guide-to-bad-data/ – Data observality can save your business or at least your nerves. Because data breaks all the time.
https://www.reddit.com/r/MachineLearning/comments/sdgqjt/d_anyone_work_at_a_bank_how_do_you_document/ – Reddit thread about documentation of predictive models for audit purposes in banks all around the world.
https://www.reddit.com/r/dataengineering/comments/schltg/beginner_mistakes_to_avoid_in_building_data/ – Another interesting Reddit thread, this one is about avoiding the beginner mistakes in data pipelines.
https://www.kdnuggets.com/2022/01/models-rarely-deployed-industrywide-failure-machine-learning-leadership.html – Only 20 % of models are deployed to production.
Video & Podcast
Papers & Books
https://www.sciencedaily.com/releases/2022/01/220126122501.htm – MonoCon is a new technique that identifies 3D objects in 2D images. Original article is here.
https://wesmckinney.com/book/ – Python for Data Analysis, 3rd edition is freely available in HTML.
https://christophm.github.io/interpretable-ml-book/ – Interpretable Machine Learning is free online.
Behind the Fence
https://j.brt.mv/jb.do?reqGK=27626912 – Data Scientist in Insperity, Houston, Texas, USA.