Radim

29 ms·

271.
by Radim 8y ago
Ah, good old "outsource non-vital parts, so you have more time for yourself (sic)". How much of your life can be outsourced, before it's no longer your life? How much of a man still makes a man? Edgar Allan Poe had a funny
272.
by Radim 8y ago
I'm afraid I can also only offer anecdata. Mine is based on hiring and then working with dozens of SW engs since 2008 (my first tech lead role), plus our student Incubator and open source projects (dozens of more junior devs). The corr
273.
by Radim 8y ago
"Implementing Poincaré Embeddings" (hyperbolic embeddings implemented in Gensim): https://rare-technologies.com/implementing-poincare-embeddin...
274.
by Radim 8y ago
Thank you! That's so nice to hear :) To get on-topic: Gensim could use such a rewrite as well… The ML world changed, expectations and requirements changed, ecosystem and APIs changed. I changed too.
275.
by Radim 8y ago
The dance between death by ossification and death by excessive chaos is a delicate one. Be too quick and you're constantly chasing shadows… wait too long and you're immobile.
276.
by Radim 8y ago
Buried in the "2.11 Frequent rewrites" section, but a great hack for "productivity via a sense of ownership": "In addition, rewriting code is a way of transferring knowledge and a sense of ownership to newer team m
277.
by Radim 8y ago
The China Study is famously biased and scientifically dodgy [0, 1]. It was great for opening the topic and provoking discussion though, which is much needed. [0] https://sciencebasedmedicine.org/the-china-study-revisited&#x2
278.
by Radim 8y ago
This is both wrong and dangerous. I do hear this sentiment often: "Hey dudes, it's just a prototype! Why waste time making it pretty? Who cares? We're doing SCIENCE here!" But prototypes (and beginnings in general) are
279.
by Radim 8y ago
> … applying semantic indexing … to the sparse data in a way that was both fast and let us store hundreds of millions of documents in memory on fairly normal hardware. Fascinating. Would you mind sharing more, some links / white pap
280.
by Radim 8y ago
Many of the errors were corrected in later editions (it's been in print for 14 years!). You can see the original errata here: http://errata.wikidot.com/0767908171 I leave it to your judgement whether this makes you &qu
281.
by Radim 8y ago
Bill Bryson's "A Short History of Nearly Everything" is an amazing layman start. It's a little dated (the first edition came out in 2004!) but to be honest, if you're interested in a "human-focused" intro
282.
by Radim 8y ago
> It was not an auspicious start to my career in computer science. I'd argue the opposite! Learning how to "run" code in your head, cover edge cases and invariants, is a great skill to practice. When I was at a hospital af
283.
by Radim 8y ago
It's not just sentiment classification. Many of these NLP methods overfit to the minutiae of the particular SOTA data set and data type. Language identification, NER, POS tagging, parsing, tokenization… you name it. Solved in theory, t
284.
by Radim 8y ago
For those who didn't catch your (brilliant and well-placed) Monty Python reference: Four Yorkshiremen , https://www.youtube.com/watch?v=26ZDB9h7BLY
285.
by Radim 8y ago
Lest you think this is a problem only for Africans & USA: I was denied visa, going to present my paper at NIPS 2010 in Vancouver, Canada. Czech citizen (EU), PhD student at the time, with obvious ties to CZ. I fully respect each country
286.
by Radim 8y ago
That's what we did, too. As a company, we formalized our open source R&D activities ( https://github.com/RaRe-Technologies ) and launched a "Student Incubator" programme. Junior people around the world get
287.
by Radim 8y ago
Mandatory link to "Idiocracy" and holding back on having children (it seemed like a joke back then!): https://www.youtube.com/watch?v=-N9nVLXMhPc
288.
by Radim 8y ago
I remember when Facebook was still figuring itself out, back in the day, users could purchase digital gimmicks directly there, to send to other users. Some "pokes" or emojis or something like that, I topped up my FB balance with l
289.
by Radim 8y ago
Yes, apparently collector's value arrives in generational waves. The first wave hits in 20-30 years, when kids grow into productive age, and sentimentally overvalue things from their childhood. A grandma's rug or ex-libris, is you
290.
by Radim 8y ago
They didn't "remove vowels" as much as started out without them! We're talking times when writing itself was utter novelty, a curiosity performed by a handful of scribes primarily for accounting purposes, and of no great
291.
by Radim 8y ago
Could be. Using Jupyter for ML development or even prototyping (as opposed to presentations / demonstration / teaching like the OP — that's where Jupyter really shines) is a red flag. I see a similar pattern with Pandas: some
292.
MIT's Lex Fridman on the State of Robotics and AGI [PODCAST](youtube.com)
2 points by Radim 8y ago|0 comments
293.
by Radim 8y ago
This is exactly correct. The difference is in how you approach problem solving, not how fast you write the code. Can you clearly articulate the important concepts and how they inter-relate? Define the problem boundaries (and later, as a con
294.
by Radim 8y ago
Right. Toppling elected governments coupled with an age-based genocide would be a wonderful solution. It's the "young and progressive" who know best! People with life experience are truly lost. /s
295.
by Radim 8y ago
This reminds me of the so-called "dark ages". Bronze age collapse, Middle ages… Early historians used to call them dark because of just such "limited evidence". Little surviving records, compared to the periods before&
296.
by Radim 8y ago
OP paywalled, but what's funny is that your comment could be read in two completely opposing ways: 1. Zuckerberg tells employees not to mistreat diverse views like Google did with Damore 2. Zuckerberg tells employees not to mistreat di
297.
by Radim 8y ago
Ha! We do offer powerful "personal information analytics" in terms of GDPR, https://pii-tools.com :) It's more for auditors and DPOs though. The challenge with GDPR is more simplicity, taming the mess. Rather than
298.
A guide to mathematical emotions(mathwithbaddrawings.com)
8 points by Radim 8y ago|1 comments
299.
by Radim 8y ago
Mandatory "How things are VS What people say" graph: https://twitter.com/RadimRehurek/status/857454177106116608
300.
by Radim 8y ago
I couldn't find any numbers (incl. your "90%"), more clarity would be appreciated, though I understand Tidelift is still an experiment. But this explanation [0] caught my eye: > How we compute how much to pay > A wei

More ›