HNHacker News
TopNewBestAskShowJobs

anjc

1,414 karma · joined October 24, 2011

submissionscomments
anjc··on Using Reinforcement Learning in the Algorithmic Trading Problem
Because they're academics who choose to forego money in order to advance their career in academia. They could also generally double their income overnight by choosing to move to industry. Many do, most choose not to.
anjc··on Using Reinforcement Learning in the Algorithmic Trading Problem
Only if you don't have infinite money on the side, in which case it's guaranteed to be profitable
anjc··on Show HN: A basketball hoop to maximize shots that go in [video]
Thanks
anjc··on Show HN: A basketball hoop to maximize shots that go in [video]
Thanks!
anjc··on A List of Colleges’ Plans for Reopening in the Fall
In my experience, the primary limit for engineering/CS labs are simply room capacities and the number of PCs per lab. Each lab therefore must be tightly scheduled by the school. Virtual learning has been amazing in this sense: labs can be rescheduled, have unlimited capacities, can run concurrently, etc.

They're also far easier to manage, less noisy, you aren't on your feet for hours.

anjc··on A List of Colleges’ Plans for Reopening in the Fall
> Colleges are not going to let many people take gap years.

How are they going to get out of abiding by their own regulations? Universities I'm familiar with all provide for deferred years.

anjc··on Show HN: A basketball hoop to maximize shots that go in [video]
Can someone briefly eli5 how the Monte Carlo simulations feed into learning the optimal surface?
anjc··on How the iPad's New Trackpad Works
Could Apple go back to being forward looking with their tablet/laptop/computer line please? If this keyboard/trackpad is a good idea now, then it would have been a good idea 10 years ago. If it was a good idea 10 years ago then they simply lacked the foresight to push it. It has also been a point of critique for as long, and I'm sure that I am not the only person who had to stop pretending that iPads were usable for (my) work and switched to PC for portable productivity (Surface Pro).

At this rate I wont be surprised to see revolutionary full access to the filesystem, and flexible window management, in 2030.

anjc··on The Longest Record Broken: Gold/Silver Ratio Hits Highest in Over 5,000 Years
Isn't this what some governments (through purchasing supply) and some central banks (through relaxed lending restrictions) have done with real estate?
anjc··on Gilbert Strang Teaches Linear Algebra
> I gained some great intuition from http://matrixmultiplication.xyz/ and (of all things) a paper named "An Introduction to Quantum Computing"[0].

Just wanted to say thanks for this paper. It really does help with intuition and it's not something I would've expected to get through.

anjc··on Dreamcast Emulator Redream 1.5.0 Progress Report
> many years ago I tracked down a working Dreamcast and acquired Sonic Adventure 2

You make it sound like a relic! There are plenty for sale and new games are still made for it

anjc··on Apple engineer killed in Tesla crash had previously complained about autopilot
I hope people aren't commonly doing what this poor guy did, turning autopilot on and taking their hands off the wheel at 71mph. At this speed, the stopping distance is greater than the max effective distance of some of the forward looking cameras, and a collision means almost certain death. I'm surprised people will trust any autopilot system at a potentially fatal speed.
anjc··on Using neural networks to upscale a famous 1896 video to 4k quality
> A problem may be that the source material is simply too high quality to begin with? I watched the original first and thought "wow, 1896, this enhancement is great", only turns out that was the original clip.

Funny, I did the same

anjc··on Using neural networks to upscale a famous 1896 video to 4k quality
Can people in this thread genuinely not see a difference between the before and after or are you just being contrary? The change in resolution should be clear enough, but the FPS change is amazing and gives a great sense of presence to me when watching.

There are some jarring artifacts, such as the ladies' hats glitching at 10sec, but these artifacts are in the original and simply become clear in the upscaled version. It's very impressive.

Side note, They Shall Not Grow Old is just stunning, with its restoration pipeline. I can't wait until all media can be updated in such a fashion, and I assume it's just a matter of time before e.g. YouTube and Spotify have filters to offer restored VHS footage, old phonograph recordings, etc

anjc··on U.S. bill to exempt small virtual currency transactions from CGT [pdf]
I'm confused then, we seem to be in agreement? If you're not a professional day trader, and you don't trade crypto as a primary source of income, then are your trades not "personal" going by the definition you gave?
anjc··on U.S. bill to exempt small virtual currency transactions from CGT [pdf]
> recommend that anybody who wants to avoid criminal charges should not base any decisions on your definition of a personal transaction.

Well of course

anjc··on U.S. bill to exempt small virtual currency transactions from CGT [pdf]
A capital gain of $200 for a single transaction is plenty for day traders who do it in their spare time. This is still a personal transaction, even if you happen to profit from it.
anjc··on U.S. bill to exempt small virtual currency transactions from CGT [pdf]
If it's not your main source of income then it's presumably a personal transaction, i.e. it's not a company transaction. This would cover most crypto day traders.

That you're buying Litecoin from a stranger with your Bitcoin doesn't make it not personal.

anjc··on U.S. bill to exempt small virtual currency transactions from CGT [pdf]
Per transaction! This seems like it'll make day trading tax free, if it isn't your main source of income
anjc··on My story as a self-taught AI researcher
You're correct, I'm sure it could happen, it is highly unlikely that a non-academic could get published even at workshops for low ranking conferences. In my experience it takes at least a year of guidance before the average PhD can be accepted to any conferences without supervision. If OP can do this alone then he's exceptional.
anjc··on Postdoc Myths
These grants tend to require industry partners and matched funds, so the value for money with regard to public funds is probably far higher now than the "weird" era you're talking about.

> enrich established PIs

Nobody stays in academia to be "rich", and in general they could earn far more and be far less stressed by working in industry.

anjc··on Postdoc Myths
The titles for these roles are not "postdoc" though, that's his point no? The role "postdoc" doesn't exist, at least in any institutions I know. Of course there are postdocs in the world, but the actual payscale that you're on may be one that a person without a PhD can be hired on, e.g. research assistant. They then call this a "postdoc" position to indicate that you need a PhD to be hired, but the contracts and conditions do not specifically cater towards people pursuing careers in academia.
anjc··on My story as a self-taught AI researcher
> I guess I was wondering where math fit in this picture for AI research.

Well you can't do #1 or #2 without having a level of maths proficiency that most college grads do not have.

FWIW, I don't understand the difference between #1 and #2 above. Most academic/industrial research is incremental (i.e. #1), and a tiiiiny % will have any impact in the way something like XGBoost would (the example he gave in another comment). That doesn't mean that the non-impactful research isn't 'basic'. You could alternatively just call #2 "groundbreaking research" and #1 "non groundbreaking research, but you need mathematics knowledge for both imo.

anjc··on My story as a self-taught AI researcher
Most people would be wrong if they think that this is what all researchers do
anjc··on My story as a self-taught AI researcher
Your tutorial is very good, but to able read even a few paragraphs you need to be proficient with linear algebra and calculus already.

> Most papers strike me as requiring a non-trivial knowledge of linear algebra

I think this is correct, if you consider college level linear algebra and an intuition for applying it to novel problems to be non-trivial knowledge

anjc··on Postdoc Myths
Oh well I understand that funds are allocated for a project. But in my experience, postdocs who are coming to the end of the funding for one project tend to be moved to another incoming one. I.e. they're essentially permanent staff anyway. I know of people who've gone on like this for 10+ years.

"Revenue generating" centres within unis have targets and projections (e.g. commercialization teams) so they DO have an idea of what funds they'll have in the medium term, even though it's on a project by project basis. I don't see how this is any different from any company that needs revenue in order to keep staff, but yet companies manage to keep staff permanently. In fact, even universities seem to be able to keep staff permanently, unless they're researchers.

anjc··on Postdoc Myths
Good article, but I must misunderstand, how do the suggestions at the bottom overcome the issues/myths listed, particularly "postdocs formally [don't] exist" which seems to be the issue at the root of all problems.

The tenuous employment status of postdocs seems to be a choice made at an institutional level, although seemingly across every institution in the world. I'm not aware of any funding stipulations that prevent institutions from giving postdocs proper employment status/permanent contracts. Sure, a new project might require a postdoc and might have funds allocated for this, but why can't unis use their already-existing projections for incoming funds and allocate jobs (i.e. permanent postdoc positions), just as private companies hire staff on the basis of cash flow/revenue projections?

anjc··on Grindr and OkCupid Spread Personal Details, Study Says
It doesn't follow that because an app is free as in beer that the app company is selling your data to make up the cost of running the app. For one thing, the likes of Plenty of Fish got to near billion $ valuation via AdSense and banner ads. I imagine people generally think this is how apps earn their money, if they were forced to think about it, especially since lots of freemium apps simply remove some banner ad when you pay.

Secondly, many companies are getting to mega valuations with free apps, via debt.

anjc··on Ask HN: What were your most impactful podcast episodes from 2019?
The Hidden Forces podcast is excellent
anjc··on Abandoned stores, empty homes: why San Francisco's boom looks like a crisis
You're looking at GDP, not budget...

If you want to compare GDP, SF is in the top 20-30 countries according to your own link

← PreviousPage 6 of 33Next →