grumbelbart2

7 ms·

91.
by grumbelbart2 11mo ago
I triggered some bug by pausing the simulation, setting the mass of one of the objects to 29.1, then resuming. The lighter objects bounced into the massive objects a few times, then all three objects were suddenly ejected with a very high v
92.
by grumbelbart2 11mo ago
If you fixed something in an open source library you use, and you don't push that upstream, you are bound to re-apply that patch with every library update you do. And today's compliance rules require you to essentially keep all li
93.
by grumbelbart2 11mo ago
> It distracts from what is actually helping which is using different functions with nicer behaving gradients, e.g., the Huber loss instead of quadratic. Fully agree. It's not the "fault" of Backprop. It does what you tell
94.
by grumbelbart2 1y ago
> backrooms horror environments True, it sounds (and looks) a lot like https://scp-wiki.wikidot.com/scp-3008
95.
by grumbelbart2 1y ago
Also, aviation is great example of how we can manage failures in complex systems and how we can track and fix more and rarer failures over time.
96.
by grumbelbart2 1y ago
They almost always try that. Save yourself the hassle, use one of the online services who will get that money for you.
97.
by grumbelbart2 1y ago
Since their energy density is still lower, it will probably take a while for them to be adapted in EVs. But their impact on energy storage to stabilize the grid, both technically and in terms of prices, can not be overstated. Cheap, safe st
98.
by grumbelbart2 1y ago
The bigger picture is though that taking swings every day also requires a bit of "luck". Some people just don't have the energy, education, mindset, or environment. It's easy to assume it would be in our control, but if
99.
by grumbelbart2 1y ago
There are some results from Scotland already, all very positive https://academic.oup.com/jnci/article/116/6/857/7577291?logi... > No cases of invasive cancer were recorded in women immunized at 1
100.
by grumbelbart2 1y ago
Sure, but the same goes for verbatim strings, or leading whitespaces in python. "Code formatting" questions usually only concern themself with those degrees of freedom that do not alter the semantic meaning of the code.
101.
by grumbelbart2 1y ago
That is only true if your path to AGI is to take models similar to current models, and feed them with tons of data. Advances in architecture and training protocols can and will easily dwarf "more data". I think that is quite obvio
102.
by grumbelbart2 1y ago
It should be solved by smoothing the image to remove high frequencies that are close to the sampling rate, to avoid aliasing effects during downsampling. The term to search for is Nyquist–Shannon sampling theorem. This is a quite well under
103.
by grumbelbart2 1y ago
LLMs add a new thread model. If trained on your data, they might very well leak some of its information in some future chat. Meta, Alphabet might not want that, but it is impossible to completely avoid with current architectures.
104.
by grumbelbart2 1y ago
I'd expect the first exploration to be done by machines, and digital transfer of the controlling instance would remove almost all drift.
105.
by grumbelbart2 1y ago
Work is already underway in that direction: https://github.com/deepseek-ai/DeepSeek-Prover-V2 but also https://deepmind.google/discover/blog/alphaevolve-a-gemini-p... Terence Tao is doing a l
106.
by grumbelbart2 1y ago
> The focus on "local" is incidental Strongly disagree with this. It is the default go-to for companies that cannot use cloud-based services for IP or regulatory reasons (think of defense contractors). Isn't that the main
107.
by grumbelbart2 1y ago
It really sounds like that, doesn't it. Better funnel those sweet Oil dollars to some elite, so the plebs work more.
108.
by grumbelbart2 1y ago
Just wondering how SQLite would ever work if it had zero control over this. Surely there must be some "flush" operation that guarantees that everthing so far is written to disk? Otherwise, any "old" block that contains d
109.
by grumbelbart2 1y ago
You're joking, but many of the code bases I saw that were produced by/with AI-support are not maintainable by any sane human. The more you go AI, the less you can turn back.
110.
by grumbelbart2 1y ago
> Writing reviews isn’t, like, a test or anything. You don’t get graded on it. So I think it is wrong to think of this tool as cheating. Except that since last year, it kind of is. It is now mandatory for some large conferences (such as
111.
by grumbelbart2 1y ago
Even the US is a good counterexample. It is a widely fragmented country, yet that worked for decades without degrading into a totalitarian state.
112.
by grumbelbart2 1y ago
It has been back for a while in the form of quantum fields.
113.
by grumbelbart2 1y ago
It's more consistent. As a rule of thumb, Bloom filters require 10 bits per element for a 1% false positive rate. Say 100 pixels changed between the frames, that 1000 bits or ~125 bytes to store the which-pixel-changed-map. Runlength e
114.
by grumbelbart2 1y ago
> What contributed more towards success in my opinion are "shortcut connections" through layers which enable more influence on early layers during learning. For those who don't know, that is the idea behind ResNet (He et a
115.
by grumbelbart2 1y ago
Do you have an example? That just sounds like a hash that is a few bits longer.
116.
by grumbelbart2 1y ago
There is of course a slight chicken-egg-thing here: If there was no (dynamic) branch prediction, we (as in compilers) would emit different code that is faster for non-predicting CPUs (and presumably slower for predicting CPUs). That would m
117.
by grumbelbart2 1y ago
In theory, sure. In practice, complex technological and political issues remain - apparent by the fact that no country has solved the issue yet. You apparently stable salt mines start leaking. Locals don't like having toxic stuff burie
118.
by grumbelbart2 1y ago
While you are fully right, that is very US-centric. Tuition is practically free in large parts of Europe, for example.
119.
by grumbelbart2 1y ago
"Missing" information is a huge red flag as well, though. If an Arbeitszeugnis only lists the dates and what the person worked on, and nothing about results or behavior, we would never invite them for an interview.
120.
by grumbelbart2 1y ago
But that's an issue organizations bring upon themselves, by defining semi-arbitrary KPIs that are used without proper interpretation. It's not directly caused by CVEs or assigned scores. It's like blaming git that it count li

More ›