While Uber Technologies Inc. and Carnegie Mellon University announced a partnership to develop autonomous car technology in February, Uber's actions earlier in the year have left Carnegie Mellon's robotics research in jeopardy:
Carnegie Mellon University is scrambling to recover after Uber Technologies Inc. poached at least 40 of its researchers and scientists earlier this year, a raid that has left one of the world's top robotics research institutions in a crisis.
Uber envisions autonomous cars that could someday replace its tens of thousands of contract drivers. With virtually no in-house capability, the San Francisco company went to the one place in the world with enough talent to build a team instantly: Carnegie Mellon's National Robotics Engineering Center.
Flush with cash after raising more than $5 billion from investors, Uber offered some scientists bonuses of hundreds of thousands of dollars and a doubling of salaries to staff the company's new tech center in Pittsburgh, according to one researcher at NREC.
The hiring spree in January and February set off alarm bells. Facing a massive drain of talent and cash, Herman Herman, the newly elevated director of the NREC, made a presentation May 6 to staff to explain the situation and seek ideas on how to stabilize the center, according to documents reviewed by The Wall Street Journal.
The short presentation at the school here laid out the issues. In all, Uber took six principal investigators and 34 engineers. The talent included NREC's director, Tony Stentz, and most of the key program directors. Before Uber's recruiting, NREC had more than 100 engineers and scientists developing technology for companies and the U.S. military.
Original Submission
Related Stories
Uber poached around 50 scientists working on self-driving car technology at Carnegie Mellon University's National Robotics Engineering Center earlier this year. Uber had been partnering with CMU to research building its own autonomous vehicles. But then it pulled from a massive venture funding war chest to hire away a lot of CMU's talent for its Uber Advanced Technologies Center.
Now, Uber appears to be publicly trying to win back the good graces of the academic community by announcing a $5.5 million gift to CMU. The money will support hiring a new robotics faculty chair and three fellowships.
"Poached around 50 scientists." CMU scientists are like the King's Deer.
Bloomberg News reported on Monday that Google is "preparing to offer its own ride-hailing service, most likely in conjunction with its long-in-development driverless car project". Google is preparing to launch its own ride-hailing service and app despite the fact that its investment arm, Google Ventures, has already invested millions in Uber, including $258 million in its largest investment deal ever, back in August 2013.
David Drummond, Google's chief legal officer and senior vice president of corporate development, joined the Uber board of directors in 2013 and has served on it ever since... Drummond has informed Uber's board of this possibility, according to a person close to the Uber board, and Uber executives have seen screenshots of what appears to be a Google ride-sharing app that is currently being used by Google employees. This person, who requested not to be named because the talks are private, said the Uber board is now weighing whether to ask Drummond to resign his position as an Uber board member.
On the same day, Uber announced that it is building a robotics research lab in Pittsburgh, PA to "kickstart autonomous taxi fleet development," and partnering with Carnegie Mellon University's Robotics Institute.
Sources tell us Uber is hiring more than fifty senior scientists from Carnegie Mellon as well as from the National Robotics Engineering Center, a CMU-affiliated research entity.
In the past, Uber CEO Travis Kalanick has said he would replace human drivers with self-driving cars. The decision to run the facility in Pittsburgh makes perfect sense, given the proximity to CMU and the potential secrecy afforded by moving research out of Silicon Valley.
Commentators have long predicted that Uber would eventually replace its fleshy drivers with robotic cars. Now Uber is one step closer to eliminating human imperfection.
Just when you thought that anime character designs couldn't get any more generic, machine learning comes to the rescue:
A collaborative team from Fudan University, Carnegie Mellon University, Tongji University and Stony Brook University have created a generative adversarial network (GAN) that can generate high-quality anime character drawings with just minimal input from humans. You can check the demonstration website and make your own anime renders by inputting some basic preferences.
You can choose hair and eye color and then decide on what accessories you might like such as glasses, hats or ribbons. The site will then generate a character for you based on your inputs. There are some things that could be improved for sure, but overall, it's a really fun application.
Researchers used a technique called DRAGAN to train the AI. The demo site was built on ReactJS. The scientists have summed up the full process in more detail in this technical report. The source code is also available from here.
Of course, designing a character is only the first step. How about a custom Gatebox holographic waifu? Or import your character into an advanced VR fantasy world powered by petaflops GPUs. Which is just a stopgap measure until you jack into the Matrix directly, allowing you to caress your loved one (as far as your brain is concerned).
Is this what happens when you poach Carnegie Mellon University's top scientists?
"Moe" (pronounced "mo-ayyy") refers to "kawaii"/cute characters.
IEEE Spectrum is running a series of articles on the DARPA Robotics Challenge (DRC), which runs this weekend (the 6th & 7th of June) and in which the finalists will be scored on a set of tasks relevant to disaster response scenarios.
There's an overview of the course and the challenges the robots will face:
The DARPA Robotics Challenge Finals is all about the course: the sequence of eight tasks that the robots are going to try to complete in 60 minutes or less. We’ll take a look at each one of the tasks, and go through all the rules that the robots are going to have to follow as they make it through to victory.
Also a background article summarises the schedule and previous Spectrum coverage.
The associated DARPA site has additional background information and updates:
The DRC is a competition of robot systems and software teams vying to develop robots capable of assisting humans in responding to natural and man-made disasters. It was designed to be extremely difficult. Participating teams, representing some of the most advanced robotics research and development organizations in the world, are collaborating and innovating on a very short timeline to develop the hardware, software, sensors, and human-machine control interfaces that will enable their robots to complete a series of challenge tasks selected by DARPA for their relevance to disaster response.
Ed Note: See our earlier coverage of Uber Poaches Robotics Scientists from Carnegie Mellon.
Original Submission
(Score: 3, Insightful) by LoRdTAW on Monday June 01 2015, @05:42PM
An old coworker once said: "The best way to ask for a raise is to quit." And it holds true. Every public sector employee I have spoken with were always offered a counter offer to stay when they looked to move on. Of course not every offer was generous or tempting enough.
If CM knew they were quitting, why did they not make a counter offer? Uber is obviously throwing a lot of money at them.
(Score: 1, Interesting) by Anonymous Coward on Monday June 01 2015, @07:23PM
Probably the same reason why Big Boy (the DJ) didn't offer his previous employer a counter offer opportunity before moving on (I'm guessing this is part of the reason, I don't actually know). He figured the previous employer should have given him a raise by now and was taking advantage of the fact that he was either mostly unwilling or unable to move on. So he moved on without providing the previous employer with such an opportunity. Figure if the previous employer didn't have the decency to provide for a raise all on their own they didn't deserve to have him as a DJ. Or he asked for a raise and the previous employer refused so he figured the previous employer didn't even deserve the opportunity.
Also sometimes people get tired of doing the same thing over and over again. They get in a rut. They want to try something new and figure perhaps their skills and talents could more socially valuable somewhere else. Not everything is all about money.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @08:16PM
(same poster)
Another possible reason could be that the employee simply doesn't like their previous employer, doesn't like another (or the other) employee(s) or the work environment and figures they can try something different.
(Score: 4, Touché) by Anne Nonymous on Tuesday June 02 2015, @01:24AM
"The best way to ask for a raise is to quit."
And vice versa.
(Score: 5, Insightful) by Dunbal on Monday June 01 2015, @05:42PM
Yeah I always find it amusing when someone complains that their employees are being "poached". How about maybe pay your staff a little better, give them projects they believe in, support, them, etc, and then they won't LET themselves be poached. No but what the capitalist (read "parasite") would rather have is some government enforced contract making sure his slav er employee stays working where he is for the shitty wage he is getting.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @07:00PM
> the capitalist (read "parasite") would rather have is some government enforced contract
I think you're confusing the capitalists with someone else. The capitalists would tell the individual to go get the best deal they could for themselves, just as happened here.
(Score: 2) by ikanreed on Monday June 01 2015, @07:41PM
No, he's referring to the Marxist concept of the "capitalist class", i.e. those whose incomes primarily come from the ownership of capital, and paying employees to make money from that capital.
The real world is way more complicated than that(In this case a university is a tool for the advancement of society, not profit, for example), but they weren't referring to those who value the philosophy of capitalism. Overloaded language is always fun.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @08:12PM
Yeah, well those must be theoretical capitalists. All the real world capitalists I know are free-markets (including labor) people.
(Score: 2) by ikanreed on Monday June 01 2015, @08:35PM
Okay, you're not getting this, and that's okay.
In usage A(the GGGP's) the word capitalist refers not to someone who holds a set of beliefs, but instead, a particular social class delineated mostly by material wealth. I was not accusing these people of being marxist, but instead being an idea that is written about in Marx's work.
In usage B(yours) the world capitalist refers to someone who values the ideals of capitalism.
I'm not sure how you'd know if you'd "met" the former group, because they aren't people rapidly pushing an ideology, but an identity group used to interpret and understand society under one long-dead academic's system.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @08:51PM
Maybe I'm an idiot, or maybe it's Marx, but only one of us is creating an imaginary group of theoretical people who want to crush the little people at the expense of all other values.
(Score: 0) by Anonymous Coward on Tuesday June 02 2015, @02:12AM
Maybe I'm an idiot,
No need to be all hypothetical about it!
(Score: 0) by Anonymous Coward on Tuesday June 02 2015, @03:19AM
Look, they call it Marxist Theory for a reason, because it's just that, a theory. Nobody has proved anything Marx said, otherwise it would be a law. So as far as I'm concerned, the guy just made the whole thing up.
(Score: 2) by ikanreed on Tuesday June 02 2015, @01:27PM
Never meant to imply otherwise, but his terminology did make it into the standard nomenclature.
(Score: 1, Informative) by Anonymous Coward on Tuesday June 02 2015, @01:31AM
All the real world capitalists I know are free-markets (including labor) people.
That's interesting, the capitalists I know about all spout free-market bs, but deep inside they all want to be monopolists.
In other words, competition gets in the way of making the really big money and/or having complete power/control...
(Score: 1, Touché) by Anonymous Coward on Monday June 01 2015, @07:47PM
No, the capitalist would coerce the individual to work for the capitalist at the cheapest possible wage. Then when hired, the capitalist would haggle with the individual to lower that even further threatening to outsource the job if the individual doesn't take the pay cut.
(Score: 3, Insightful) by GungnirSniper on Monday June 01 2015, @07:40PM
May Ayn Rand bless you!
Tips for better submissions to help our site grow. [soylentnews.org]
(Score: 0) by Anonymous Coward on Monday June 01 2015, @06:10PM
If you don't give a good enough package of money/benefits/prestige/environment/interesting work, people will move on to competitors. If you want to avoid it, make some or all of those things better. It's not a hard thing to understand.
And if you don't offer a strong enough package, your people leave and you die. You don't have a guarantee to exist.
(Score: -1, Troll) by Anonymous Coward on Monday June 01 2015, @06:29PM
"Free market", bitches!
Now, don't the dullards at English dept feel like crap now? :)
(Score: 1) by MyLongNickName on Monday June 01 2015, @06:37PM
If memory serves, this is how Microsoft put Borland into obscurity. While I am all for the worker making every buck he can, at a certain point it becomes anti-competitive.
(Score: 1, Insightful) by Anonymous Coward on Monday June 01 2015, @06:43PM
Let's see what happens when the Uber bubble pops. R&D is always the first thing on the chopping block.
(Score: 2) by mhajicek on Tuesday June 02 2015, @02:22AM
Hopefully those researchers and engineers are able to put some of that money in savings to cover the employment gap, then use their experience at Uber to get another good position.
The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
(Score: 3, Funny) by CoolHand on Monday June 01 2015, @06:53PM
and the next day the rest of the staff left for Uber after hearing of the big bonuses and doubled salaries...
Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
(Score: 5, Insightful) by PizzaRollPlinkett on Monday June 01 2015, @07:47PM
Wait until Uber folds, and they'll be back. Can't blame them for spending a few years making $ while they can. CMU ought to consider this a sabbatical or something. Uber can't possibly make self-driving cars before their business model implodes. Uber subtracts value from the marketplace by skimming money from transactions and not paying enough for people to maintain their own vehicles. Things like Uber are a cancer in the economic body. They can't possibly replace drivers with automation before they burn through all their money, even with the CMU brain trust.
Links:
http://www.buzzfeed.com/johanabhuiyan/what-uber-drivers-really-make-according-to-their-pay-stubs [buzzfeed.com]
http://www.washingtonpost.com/news/get-there/wp/2015/02/20/the-hidden-costs-of-being-an-uber-driver/ [washingtonpost.com]
(E-mail me if you want a pizza roll!)
(Score: 3, Interesting) by isostatic on Monday June 01 2015, @09:38PM
It's supply and demand. If the drivers don't make enough money, they won't drive, and prices will have to go up. Until then, invisible hand FTW!
However I don't travel via uber because of the price - I travel because the money goes off via credit card and I get a statement at the end of the month. No silly tipping, no drivers wanting cash "oh the card machine broke", and no fighting for receipts after a 16 hour day. On top of that the cars are always cleaner than a taxi or minicab, the drivers more professional, and you don't feel you're a burden to them.
Instead of standing in the rain trying to hail a cab, I get my phone out, book it, see exactly when it arrives, get in, travel, and leave. Every month I print off my statement and post it in with my expenses - job done.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @11:51PM
It's supply and demand. If the drivers don't make enough money, they won't drive, and prices will have to go up. Until then, invisible hand FTW!
Those wages won't be going up with a large pool of unemployed and under-employed people in society to compete with.
(Score: 2) by isostatic on Tuesday June 02 2015, @12:07AM
So supply is high, and likely to remain high. The taxi lobby have generally failed (a few places they've won for now - like Germany). Where's the uber implosion going to come from? A competitor that can do the same thing but with a lower margin? Uber has enough cash to loss lead for a long time in markets where's there's like for like competition.
Ubers biggest risk is someone beating them to market with robotic vats, but if drivers are only netting $12 an hour, there's still room to be competitive for a long time even if tesla or Google gets an automatic car up and running. Think the hurdles for uber are bad, that will be nothing compared with johnnycab.
(Score: 1, Interesting) by Anonymous Coward on Monday June 01 2015, @08:52PM
Shades of
http://www.mercurynews.com/business/ci_7044112 [mercurynews.com]
(Score: 1) by shipofgold on Monday June 01 2015, @09:29PM
As far as I can tell, Uber is taking the Amazon strategy as outlined by:
http://www.joelonsoftware.com/articles/fog0000000056.html [joelonsoftware.com]
They are doing exactly what they should be doing...with other people's money.
(Score: 0) by Anonymous Coward on Monday June 01 2015, @09:53PM
Don't react based on the initial adrenaline rush, i.e. cutting off access to Uber and sending in the lawyers.
This might actually be a net positive for CMU and its robotics program... you guys need to sit and think about it.
(Score: 2) by MichaelDavidCrawford on Monday June 01 2015, @10:13PM
I expect robotics might have more job security than either physics or biology. I don't really know. But among the reasons I haven't completed my physics doctorate is that I have few illusions about a phd enabling me to get a research post at a university.
It's not just me; it is quite difficult for anyone to get a professorship. It's hard even to get a postdoc, but if you do too many postdocs than you are untouchable.
I'd like to do research but I don't regard a PhD as necessary for that. To the extent I pursue it I'll support my physics habit by coding.
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by kaszz on Monday June 01 2015, @11:26PM
Just make fusion power a working solution and your career as a professional juice drinker at a blue beach is secured ;-)
(Score: 2) by MichaelDavidCrawford on Monday June 01 2015, @11:30PM
there are many reasons for Big Science but there is still quite a lot of physics one could do in one's garage.
For theoretical physics all one REALLY requires is pencil and paper.
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by kaszz on Monday June 01 2015, @11:31PM
And someone that finances the time spent using that pencil and paper.
(Reminds me of that pencil and paper has something about it. It's so "free" in some way.)
(Score: 2) by MichaelDavidCrawford on Tuesday June 02 2015, @02:45AM
in reality, he was a magistrate of the french court.
That scientists are expected to be professors is a recent phenomenon.
Yes I Have No Bananas. [gofundme.com]
(Score: 0) by Anonymous Coward on Tuesday June 02 2015, @04:19AM
I asked this on slashdot the other day but didn't get a satisfactory answer. Perhaps I asked it incorrectly. What I was getting at is: What were the thoughts on the ability of the standard model to calculate probabilities greater than one before the Higgs solution was suggested? What is the history behind this problem?
(Score: 2) by MichaelDavidCrawford on Tuesday June 02 2015, @04:28AM
admittedly I should, I apologize that I don't.
I had just started quantum field theory in grad school when I became profoundly psychotic, admitted myself into a psych hospital, then upon my discharge dropped out of school.
I'd like to go back sometime but I'm not so sure I'll do particle physics. There are many reasons I would enjoy the work but I want to solve more pressing problems, like how to feed hungry people.
Yes I Have No Bananas. [gofundme.com]
(Score: 0) by Anonymous Coward on Tuesday June 02 2015, @06:05AM
I'm pretty sure grad school drives a certain type of person crazy.
(Score: 2) by MichaelDavidCrawford on Wednesday June 03 2015, @02:14AM
as well as the dungeon-like environment of my office and the computing lab.
I often speculate that a hundred dollars worth of tasteful lighting would have earned me my doctorate.
Yes I Have No Bananas. [gofundme.com]
(Score: 0) by Anonymous Coward on Wednesday June 03 2015, @12:42AM
I thought that feeding hungry people was all about distribution (logistics) -- the food is available, just not getting to all the places it's needed? In some cases, local despots may be grabbing the food for themselves, or to starve out their enemies?
(Score: 2) by MichaelDavidCrawford on Wednesday June 03 2015, @02:10AM
we have lots of food but it's not equitably distributed.
Why just today I read in The Columbian of Vancouver, Washington that yellowfin sole trawlers throw back 82,000,000 pounds of halibut each year, because one is required to catch halibut with hooks and lines whereas yellowfin is caught with nets. Why don't they feed that fish to the poor?
In the developing world a very common problem is official corruption. That's not just the case for foreign aid; India has always struggled to feed its own poor but the vast majority of the money or food or what have you is taken by corrupt officials.
There are many approaches one might take. I'm not completely clear how I could do it with a physics phd - if I were so clear, then I'd already be in grad school.
The US Agency for International Development contracted with a former employer of mine - "Geonex, the Nation's Largest Mapping Company" - to perform aerial photography of the Nile River so as to improve crop production there. It's hard to explain in a quick comment but there are lots of things that multispectral remote sensing can do to improve agriculture. Among the benefits of this is that Geonex was working directly with the farmers, rather than through the government or corporate business structure.
Consider that just recently someone invented a novel one-wheel plastic tank, in which the tank is also the wheel, one can fill it with ten or twenty gallons of water then pull it by a handle. For millenia, women have traditionally carried pots of water on their heads.
There are ways this can go terribly wrong. A massive foreign aid project dug hand-pumped water wells all over Bangladesh. In the short term this was unpopular as the women regarded carrying water as a way to get away from the men. In the long term, the water was found to be laced with arsenic.
Yes I Have No Bananas. [gofundme.com]
(Score: 2) by kaszz on Monday June 01 2015, @11:38PM
It would be interesting to read about how the NREC center will manage this situation. One aspect is if this will affect the salary of those that remain? Where will they get new talent? Will their reputation level be degraded? Does students courses take a hit?
Interesting case study!
(Score: 3, Funny) by rts008 on Tuesday June 02 2015, @01:32AM
How do you prepare the scientists for poaching?
How are they served, and what sauces/side dishes go best with them?
What type of scientists are best for poaching?
And most importantly, which wine should accompany the poached scientist?
So many questions...
(Score: 0) by Anonymous Coward on Wednesday June 03 2015, @12:52AM
* How do you prepare the scientists for poaching?
Butter them up by offering higher pay! (unspoken, in exchange for their soul).
* How are they served, and what sauces/side dishes go best with them?
With a contract that includes secrecy and non-compete agreements, as well as rights to any patents.
* What type of scientists are best for poaching?
The ones that have already solved the problems the company is looking to commercialize.
* And most importantly, which wine should accompany the poached scientist?
That will come from the university or other former employer...