Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dollo_7
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
dollo_7
6mo ago
Not sure if I buy it. First, SVD decomposition to obtain U, Σ, V is computationally expensive, so it would work only if we are not finetuning very big models. But my real concern comes at the results. The "13 parameters" looks lik
2.
▲
by
dollo_7
1y ago
Very interesting. We will see soon a rise of training assistants reading through our wearable sensors. Sadly, it seems these foundation models are still not open to the public. I can't find any links within the research page or the pap
3.
▲
by
dollo_7
2y ago
It is also about trying to get the most of that hypothesis testing, defining success and failure the best you can. I have encountered this "mediocre success" many times in AI solutions due to lack of problem definition. For instan
4.
▲
by
dollo_7
2y ago
I'm not sure if that is a metric you can rely on. LLMs are very sensitive to the position of your item lists along the context, paying extra attention at the beginning and the end of those list. See the listwise approach at "Large
5.
▲
by
dollo_7
2y ago
I hoped it was too good to be just a joke. Still, I will try it on my eval set…
6.
▲
The Case for Selective Slackerism
(theatlantic.com)
1 points
by
dollo_7
3y ago
|
1 comments
7.
▲
by
dollo_7
3y ago
Without paywall: https://12ft.io/proxy?q=https%3A%2F%2Fwww.theatlantic.com%2F...
8.
▲
by
dollo_7
3y ago
Academic silo with little to no real transfer to business. ML eventually enabled building better continuous and discrete models for inference, control, and prediction.
9.
▲
by
dollo_7
3y ago
I have sometimes found myself into that situation. I have to be careful not to overthink where to put my "brain room" for that day, otherwise I carry this overhead burden that rumbles all day long, questioning if I should be putti
10.
▲
by
dollo_7
4y ago
I totally agree. I work on the private sector, coming from a research position too. I was also focused on the "interesting" side of the problem: the modeling, integrating domain knowledge into the analysis, drawing all sorts of pl
11.
▲
by
dollo_7
4y ago
I believe the article is oriented to those who feel like the "nice guy" the author is describing, which is my case, and expects you to act on how you feel to his words. I feel myself very reflected on the profile seeking for high
12.
▲
by
dollo_7
4y ago
While you have a point, this process may make it harder to be a troll. Nowadays you can be a troll anywhere and anytime you want, specially online. I'm curious how this approach may benefit those who share their plain thoughts without
13.
▲
by
dollo_7
4y ago
The author's point of view seems to me quite condescending. Like being the janitor is considered less (and that, inevitably, everybody sees the janitor that way), so the author is trying to flip that view around (because he can, he is
14.
▲
by
dollo_7
4y ago
I also found writing useful to stop my head from circling around problems repetitively. Writing serves me not only to think, but to actually see the structure of that thinking process (a beginning and an end, argumentation), hence seeing th
15.
▲
by
dollo_7
4y ago
I may belong to that silent majority. I have learnt a lot from the discussions here in HN and other sites, but I rarely participate on them. This post made me think about being a little more active because, sincerely, most of the times it i