jampekka

10 ms·

241.
by jampekka 10mo ago
Nobel prizes in physics are awarded typically with lag of 20-30 years. In early 2000s China was still a relative backwater economically (and academically). In 2000 US R&D spending was over 8 times China's. Now China has likely surp
242.
by jampekka 10mo ago
The World Happiness Report discusses this: "The large variations in the systems and processes to define mortality causes imply there may be very different numbers of deaths that are registered with a specific cause. This creates a prob
243.
by jampekka 10mo ago
The World Happiness Report extensively discusses positive and negative affect in Chapter 2 and the relatively high suicide/death of despair rates of the Nordic countries in Chapter 6. These seem to be totally ignored in TFA. https:&#x
244.
by jampekka 11mo ago
LLMs can interact with the world via e.g. function calling.
245.
by jampekka 11mo ago
A good heuristic is that if an argument resorts to "actually not doing <something complex sounding>" or "just doing <something simple sounding>" etc, it is not a rigorous argument.
246.
by jampekka 11mo ago
That seems somewhat similar to perplexity based detection, although you can just get the probabilities of each token instead of picking n-best, and you don't have to generate. It kinda works, but is not very reliable and is quite sensi
247.
by jampekka 11mo ago
These figures seem to include ownership of mutual funds. https://fred.stlouisfed.org/series/WFRBST01122
248.
by jampekka 11mo ago
1491 vs 1418 ELO means the stronger model wins about 60% of the time.
249.
by jampekka 11mo ago
> Have you tried Polars? It really discourages the inefficient creation of intermediate boolean arrays such as in the code that you are showing. The problem is not usually inefficiency, but syntactic noise. Polars does remove that in som
250.
by jampekka 11mo ago
> And pandas is essentially a separate programming language. I'd say dplyr/tidyverse is a lot more a separate programming language to R than pandas is to Python.
251.
by jampekka 11mo ago
I wonder what the last example of "logistics without libraries" would look like in R. Based on my experience of having to do "low-level" R, it's gonna be a true horror show. In R it's often that things for whic
252.
by jampekka 11mo ago
https://docs.ultralytics.com/tasks/segment/
253.
by jampekka 11mo ago
Segmentation doesn't need to count legs. I'd guess something like YOLO could segment 5 legged dogs too.
254.
by jampekka 11mo ago
> over a 100€ out-of-pocket My understanding is that you typically pay something like this in the US for a specialist visit even if you have insurance, especially if you haven't already paid the year's deductibles.
255.
by jampekka 11mo ago
You can make a PWA fully offline with someting like a 10 line service worker.
256.
by jampekka 11mo ago
I also use tools like reveal.js or slidev for lecture slides now and then, and almost always come to regret it. It easily explodes into cascades of hacks if you need anything but the basic layouts offered out the box. Quite basic things lik
257.
by jampekka 11mo ago
At least in the mathematically simpler scenario of a gaussian prior and gaussian observations, the posterior mean is computed by weighing by the the inverses of variances (aka precisions) just like this. https://en.wikipedia.org&
258.
Zero-Click Remote Code Execution in Android System(github.com)
3 points by jampekka 1y ago|0 comments
259.
by jampekka 1y ago
Grok has plenty of censoring. E.g. "I'm sorry, but I cannot provide instructions on how to synthesize α-PVP (alpha-pyrrolidinopentiophenone, also known as flakka or gravel), as it is a highly dangerous Schedule I controlled substa
260.
by jampekka 1y ago
At least The Guardian has a comment from an independent expert: "Prof Carlos Frenk, a cosmologist at the University of Durham, who was not involved in the latest work, said the findings were worthy of attention. “It’s definitely intere
261.
by jampekka 1y ago
What does that mean? That USA is somehow killing people all around the world as a puppet or Iceland or something?
262.
by jampekka 1y ago
In general only combatants are allowed to be targeted. (Alleged) drug trafficking is not combatting. But in this case the point is a bit moot anyway as laws of war apply only to losers.
263.
by jampekka 1y ago
I wouldn't be surprised if autoregressive LLMs had some chaotic attractors if you stretch the concept to finite discrete state (tokens).
264.
by jampekka 1y ago
It was a joke referring to his essay. https://en.wikipedia.org/wiki/Bitter_lesson
265.
by jampekka 1y ago
Sutton still hasn't learned his own Bitter Lesson? ;)
266.
by jampekka 1y ago
Just run `pnpm install` after switching the branch. Or add `pnpm install` into the build step. And many build tools will do that automatically. If the deps are in sync with the manifest, that takes typically less than a second. This is a pr
267.
by jampekka 1y ago
For complex things there's CLI. For even more complex things there are programming languages.
268.
by jampekka 1y ago
> Well, the outcry is completely justified. Suppose a video conversion app really did have just have a drop-target area and a "do it" button. It would be ridiculously bad. That kind of crutch is ok to install for illiterate use
269.
by jampekka 1y ago
Where do they tell to use keyboard shortcuts? I've been using Gnome 3 since it came out and I haven't encountered situations where I could do things with keyboard I couldn't do easily with mouse.
270.
by jampekka 1y ago
I don't follow. In pnpm there's a global cache at ~/.pnpm with versioned packages and node_modules has symlinks to those. Dependencies are defined in package.json transitive dependencies are versioned and SHA512-hashed in pnp

More ›