HN user
In the same vein, Asimov in 1956:
Baley shrugged. He would never teach himself to avoid asking useless questions. The robots knew. Period. It occurred to him that, to handle robots with true efficiency, one must needs be expert, a sort of roboticist. How well did the average Solarian do, he wondered?
You come off as snarky, but I kind of agree. We tried this first.
It turns out digital collaborative calendars are pretty great for us in general, there is no chance in hell I could keep the analog one up to date, so it was definitely worth having a screen on the wall.
To answer the q above, this is what we have, too.
Second this.
I ended up building a nice charging station right near the entrance. It has storage for keys, wallet, and other things to grab when heading out. It has an abundance of wired and wireless chargers for all devices.
Then I got a dumb (but nice) alarm clock for the bedroom.
Then I noticed that a common reason to pick up the phone is to check the calendar. I ended up hanging a monitor on the wall, displaying the family month/agenda calendars. It’s read only, but it prevents a lot of device checking.
Cannot recommend enough restructuring physical reality to not have device on your person at home. It also helps the kids to put theirs away and learn good habits.
I suppose the Murderbot Diaries is also on the theme https://marthawells.com/murderbot1.htm
Relevant book rec: Dogs of War by Adrian Tchaikovsky. https://adriantchaikovsky.com/dogs-of-war-series.html
I met Keats through Dan Simmons - Hyperion. It's one of those books that hasn't quite left my mind despite finishing it some time ago.
Thank you!
“Hard” meaning specifically how many lawyer/HR hours, how long are the timelines, and are there other significant risks?
Are there good ways for US companies to host international grad students in their internship programs? How hard is it for them?
Specifically I had a conversation with a few administrators at universities in Sweden and this question came up on behalf of their students.
Thanks for regularly showing up here. I went EB2NIW some years ago and still get ptsd reading these threads. Hang in there everyone struggling with immigration.
I’d be interested in your thoughts on the case where the f_i are optimizable: f_i(t) = K(t, z_i), i=1..m << N. Like the representer thm but much fewer terms than you have data points to fit. The points z are usually called inducing points and may be optimized by gradient descent.
There is literature on approximating exact GP inference with (something like) these objects when m << N (variational inference).
However, I’m not aware of anyone drawing a clear picture of the other direction, starting from the optimization picture and explaining it in terms of inference, similar to what TFA does.
In TFA the number of functions is large, so the system is underdetermined. In the variational inference the system is overdetermined and I wonder what inference, if any, gradient descent does..
Caveat: 1am and a few drinks deep so if I’m not making sense that’s ok
This seems like a young talent that we’ll see more of. I like your to the point writing style and obvious passion for mathematical clarity. Keep it up and best wishes for your phd studies.
“in order to settle any lingering unease about using such tools in physics” spoken like a true mathematician :D
I enjoyed the post a lot (at least the parts that didn’t pass right over my head). But I never met a physicist with lingering unease about dimensional analysis. We get that beaten into us until it’s as natural as breathing.
You can buy this stuff in the supermarket (in Viking countries), look for paltbröd.
I use a split keyboard with custom layers and find it fine to use a normal keyboard such as when on the laptop directly. They seem to activate different modes in my brain somehow.
I have much bigger problems switching between dvorak and qwerty on either keyboard, to the point where I'm considering properly relearning qwerty despite being fluent in dvorak over a decade.
Learn to communicate efficiently and directly in writing.
I personally like McCaskill's handbook from NASA Langley because it's focused on practical matters:
https://ntrs.nasa.gov/api/citations/19900017394/downloads/19...
"Effective writing involves far more than following rules of grammar. There is a craft to creating phrases, sentences, and paragraphs that ensure communication."
Agreed. Moreover, the physical size of the device will grow exponentially with problem size, so manufacturing will be impractical. The device is small only for simple instances of the SSP. But for those instances you could simulate the device in a few lines of code on a classical computer :)
In fact, this device is based on earlier work on a biological computer [1] which suffers from the same drawbacks [2].
[1] https://www.pnas.org/content/113/10/2591.short (discussed https://news.ycombinator.com/item?id=11186640 )
[2] https://www.pnas.org/content/113/23/E3187
EDIT: it is _really_ neat engineering though. EDIT2: Old HN link & grammar
Norwegian news: https://www.nrk.no/mr/joachim-ronneberg-dode-i-natt-1.142576...
2015 nytimes interview: https://www.nytimes.com/2015/11/21/world/europe/wwii-hero-cr...
I used tuplespaces to implement a poor man's distributed computing for Matlab back in 2009 or so.
It would simply put Matlab code and parameters in a tuple, a worker would pick it up, compute, and put the results back. Used it to distribute the heavy function evaluation in a genetic optimization.
It was very easy and trouble free..
Edit: mixed up the implementations, I used TSpaces by IBM http://www.almaden.ibm.com/cs/TSpaces/Version3/ClientProgrGu...
TLDR; Recognize impostor syndrome and include 'basics' even in presentations for highly skilled people.
I'm in the physical sciences and routinely give fairly math-heavy presentations.
It's easy to think that an audience of Phd's and professors know all prerequisites and just need the hard facts in a rapid succession.
On the contrary over time I found that my audience typically appreciates when I start from the very beginning (in my case meaning advanced undergraduate level or so), and working up to the actual problem setting.
Although my audience is very much expected to know that material they A) get primed for the topic and have a chance to feel good about 'I know this' and B) understand my view of the topic so that we have a common language.
With that I'm ready to frame the technical problem and explain my work.
The last time I did this I actually spent 10 out of 20 minutes on the first part, and the talk was very well received and generated an 'invited paper' request for a journal.
No one has ever complained that my talks are 'too basic'.
EDIT: Btw, the metric I use to measure success is amount of generated questions & engagements in and after the talk. If the moderator has to cut off Q&A for the next speaker you win. If people hunt you down to talk more you win.
I agree with you that the estimate is conservative, and would depend strongly on image resolution and how broad your distribution of bedrooms is - only modern US style, or also 40 year old Japanese houses?
It has very much been fitted. The supplementary material describes the training set as 5 pictures each of 2M different scenes of the type you can see in the paper (square room with random objects).
So to extrapolate wildly, it seems reasonable that getting similar results for, say, real-world bedrooms, you'd need to take around 5 pics each of 2M bedrooms, and record the location and angle of the camera for each picture.
Edit: and I didn't mean to sound negative, using artificially generated rooms to develop the method is a great idea, and the next step will be narrow domain specific applications (they mention eg robotic arms) where it's feasible to automatically collect enough data for a task, and somewhere in the future we may have the data and compute to sample the distribution of real world environments in decent resolution..
Opeth Bloodbath Leprous Soen Haken
Mostly prog metal that I know well already. I find it hard to focus with music I have not heard before, but can quickly go into very deep focus mode with something well known.
I have an uncle who's about to retire after a career as a cobol dev. He tells me that he can teach me the language in two days, show me around the legacy systems of [big company you know] in two months, and then I could have a safe and lucrative career until I retire, too.
I wish he was interested in writing a blog post of war stories.
The 'framework build and venv on osx' thing ruined my night yesterday so this post had great timing.
Seriously, starting from a clean osx, how do you install python? I always liked using vex for venv management. But I would also like to have anaconda for some tasks..
Today you learned: Borg means fortress in swedish.
https://www.buymeacoffee.com/Jonase
Cool, thanks a lot!
Sign up was smooth on Android chrome, but I managed to misunderstand the save button once in the edit page. I pressed save and went to hamburger menu-your page, and all changes were lost. Turns out url I chose first was taken, but the error message upon saving was off screen for me so I did not see it..
No biggie, very friendly interface. I like that I got my initials on the logo cup :) buy whoever made that a coffee..
This is really cool :)
Thought, look into nanocellulose fibers for textiles, and your product could be biodegradable while still exceptionally strong[1]. Tech is probably not mature, but worth keeping an eye on for future.