dandelionv1bes

5 ms·

1.
by dandelionv1bes 8mo ago
If you’re interested in an easy explainer for backprop, then highly recommend math for deep learning by Kneusel. Finished it recently and you can apply the chain rule backwards by hand with a tiny NN, and code one too.
2.
by dandelionv1bes 8mo ago
Saw Empire of AI in a bookshop recently but held off buying as wasn’t sure if it was going to be surface level. You’d recommend?
3.
AI-exposed jobs deteriorated before ChatGPT(arxiv.org)
2 points by dandelionv1bes 8mo ago|0 comments
4.
AI–AI bias: LLMs favor communications generated by large language models(pnas.org)
2 points by dandelionv1bes 9mo ago|0 comments
5.
by dandelionv1bes 9mo ago
I struggle with this a bit because while my network isn’t bad - I really can’t stand social network like threads/X etc - I’m not on any social media bar here and LinkedIn. Do you think investing in bluesky is worth it? I’m in industry
6.
by dandelionv1bes 9mo ago
Nice read - I can’t seem to subscribe? Getting an error with Buttondown.
7.
Show HN: Tiny Toy Network – a neural net to practice backpropagation(hollyemblem.github.io)
1 points by dandelionv1bes 9mo ago|0 comments
8.
by dandelionv1bes 9mo ago
i’m sorry this really made me laugh - I hope she’s ok!
9.
by dandelionv1bes 9mo ago
I think the docs are great to read, but implementing was a completely different story for me, ie, the Ask AI recommended solution for implementing Claude just didn’t work for me. They do have GitHub discussions where you can raise things, b
10.
by dandelionv1bes 9mo ago
I do understand why it’s a product - it feels a bit like what databricks has with model artifacts. Ie having a repo of prompts so you can track performance changes against is good. Especially if say you have users other than engineers touch
11.
by dandelionv1bes 9mo ago
I’ve been slowing crunching through Math for Deep Learning, so spent a fair amount of time looking at Hessian matrices + second order optimisation. I’ve been slowly reading this book for a year, so stopping to do most of the math by hand ea
12.
Show HN: Llama 4 Maverick explores Japantown in Watch Dogs 2(youtube.com)
2 points by dandelionv1bes 9mo ago|0 comments
13.
AgentRoam: Watch GPT-5.2 control movement, camera and selfies in Watch Dogs 2(youtube.com)
2 points by dandelionv1bes 9mo ago|0 comments
14.
by dandelionv1bes 9mo ago
Amazing! Thanks will join.
15.
by dandelionv1bes 9mo ago
This is fantastic. I think it’s nailed in the substack what was missing from a lot of these LLM driven NPCs that did not feel authentic. I have a couple of follow-up questions on specifics relating to analysis of behaviour with LLMs (in gam
16.
by dandelionv1bes 9mo ago
GAMM in Rome is quite fun (Italy): https://www.gammuseum.com/en/ Shout-out to the museum of Ancient Greek technology, with their wine automaton (Athens, Greece): https://kotsanas.com/
17.
by dandelionv1bes 9mo ago
Bletchley is great! I need to go again
18.
by dandelionv1bes 9mo ago
Haven’t seen Wiby before but it’s fantastic, thank you.
19.
by dandelionv1bes 9mo ago
The response to the Sal Khan op-ed resonated with me, along with other parts of this article. Something I’ve been digging more into is some of the figures around proposed job losses from AI. I think I even posted a simulation paper last wee
20.
by dandelionv1bes 9mo ago
Thank you for putting it together. I looked at a couple of the references and they look like they point to your blog. Do you have a view at all of popular patterns in terms of citations? Might be useful
21.
by dandelionv1bes 9mo ago
I thought Gas Town was a satire until I saw the GitHub. Maybe it’s a very involved satire? It is right? “ Do not use Gas Town.”
22.
Exploring the Untouched Abandoned IBM Corporate Retreat(youtube.com)
1 points by dandelionv1bes 9mo ago|0 comments
23.
GenAI may create a socioeconomic tipping point through labour displacement(nature.com)
5 points by dandelionv1bes 9mo ago|0 comments
24.
by dandelionv1bes 9mo ago
Catch 22 is just fantastic isn’t it. I might re-read in 2026!
25.
by dandelionv1bes 10mo ago
Something I’ve been thinking about is how as end stage users (eg building our own “thing” on top of an LLM) we can broadly verify it’s doing what we need without benchmarks. Does a set of custom evals built out over time solve this? Is ther
26.
by dandelionv1bes 10mo ago
Completely agree. I was working on something with TensorRT LLM and threw Nemotron in there more on a whim. It completely mopped the floor with other models for my task (text style transfer), following joint moderation with another LLM &
27.
by dandelionv1bes 10mo ago
Complete tangent but “in lockstep” really seems to be doing the rounds again as corpo speak. It honestly annoys me but I can’t explain why - maybe because they’re not a military but an ad production company.
28.
by dandelionv1bes 10mo ago
Amazing, thank you! Will go through these.
29.
by dandelionv1bes 10mo ago
Season of the Witch is honestly such a great book. Nice to see it mentioned here.
30.
by dandelionv1bes 10mo ago
Wow, thanks for link to Paged Out. Any ideas how to discover any more tech focused zines? We have a zine culture in UK but afaik it’s more culture / music focused (v happy to be wrong here)

More ›