Live data from Hacker News

Viewing profile — nborwankar

nborwankar

HN member
Joined
Wed, Jul 14, 2010, 11:24 PM UTC
HN karma
498
Public activity
132 items

About nborwankar

No profile information was provided.

Recent public activity

  1. comment
    Comment #42301453

    Not sure if this article provides enough definitive info - happened to see it yesterday. https://www.newscientist.com/article/mg26335104-500-the-brai...

  2. comment
    Comment #41565008

    Seems ironic to be working for a company without integrity, drawing a salary ie your livelihood from such an organization while believing somehow your personal integrity is not alr…

  3. comment
    Comment #40973948

    Can you share what kinds of problems were conducive to hyperbolic embeddings in your experience. Also, separately, are you saying companies are using these in practice but don’t ta…

  4. comment
    Comment #40927773

    They have vacuum-cup like pads on their feet?

  5. comment
    Comment #40833966

    +1 for Mathematica’s documentation system. It is the best of any software doc, OSS/free or closed/commercial. By far.

  6. comment
    Comment #40710637

    This is related to a classic relational database problem called “bill of materials problem” - suggest you Google that. Also look for book on Graphs and networks in SQL by Joe Celko…

  7. comment
    Comment #40220527

    Gives a whole new meaning to “manufactured consent“.

  8. comment
    Comment #39965038

    If it could support the pgvector extension it would be a super fast vector database with all the power of Pg - the relational aspect brings the ability to add and query using rich …

  9. comment
    Comment #39946718

    +1 on fastmail

  10. comment
    Comment #39890311

    I didn’t take a year fully off but last year was slow in my consulting work so I did a deep dive on the emerging LLM area. One recommendation - get a beefy Mac laptop so you can ru…

  11. comment
    Comment #39858405

    Slightly unconnected but I’ve found a rather disturbing “dark pattern” in Amazon sales of food items in CA. Background: CA requires a warning on food that has some threshold level …

  12. comment
    Comment #39833799

    Programmers Introduction to Mathematics by Jeremy Kun should get you started. Look on Amazon. Then you can decide if you want to do a formal course and which one. I would also stro…

  13. comment
    Comment #39805340

    Here’s another one - it’s older but has some interesting charts and graphs. https://arxiv.org/abs/2303.18223

  14. comment
    Comment #39803267

    That’s not quite correct. Private data especially in the enterprise cannot use public LLM’s like GPT-4 or 5 or N. Use cases needing data privacy have to use an internally implement…

  15. comment
    Comment #39803176

    See Large Language Models - a Survey. Minaee et al https://arxiv.org/abs/2402.06196

  16. comment
    Comment #39758655

    If the robots are cloud connected we are doomed the robot apocalypse could be triggered by cyber warfare. Or by malicious human owners. If they are autonomous there better be an em…

  17. comment
    Comment #39671710

    The secular core of Buddhist thought has parallels in the Sankhya system of thought in the set of knowledge systems collectively called Hindu metaphysics. Buddhism and Hinduism had…

  18. comment
    Comment #39599663

    The question "Can you hear the shape of a drum?" was first posed by mathematician Mark Kac in a paper published in 1966. It was proven in the 80’s that there are manifolds for whic…

  19. comment
    Comment #39545699

    “Integer arithmetic is all you need” ? NVIDIA stock arrow up or down?

  20. comment
    Comment #39388107

    Re RAG aren’t you ignoring the fact that no one wants to put confidential company data into such LLM’s. Private RAG infrastructure remains a need for the same reason that privacy o…

  21. comment
    Comment #37612509

    You do get some money back for a trade-in depending on how old the machine is.

  22. comment
    Comment #37489646

    Llama and GPT are auto-regressive decoder only architectures which for pure translation jobs are not the optimal architectures. Training seq2seq models or encoder/decoder models on…

  23. comment
    Comment #37482508

    Got 502 Bad Gateway

  24. comment
    Comment #37387999

    Here’s a thread from biology stackexchange about this which also refers to CETI. https://biology.stackexchange.com/questions/108499/can-whale... .

  25. comment
    Comment #37364380

    Meta comment about the comments saying “I didn’t find what I was looking for” - core fact is it was using Wikipedia as a knowledge base. So if your topic is not represented well en…