Udacity spins out self-driving taxi startup Voyage

Image jpg

UPDATE: Sebastian Thrun answers student questions for 24 minutes in this video Self-Driving Car Nanodegree: Q&A with Sebastian Thrun. This video is probably the most informative insider perspective on the fast-moving autonomous vehicle space.

One example of how fast the field of AI is moving: Udacity’s “school for robo-cars has been so successful that it’s now spinning out of Udacity into its own company, Voyage.” Here’s a snippet from Business Insider:

(…snip…) The new spin-out will be lead by Oliver Cameron, a Udacity VP that was spearheading a lot of its self-driving car curriculum. The company broke the news to its employees Wednesday morning.

Udacity will have a stake in the newly-formed company as part of the deal, said the Udacity’s CMO Shernaz Daver. Voyage also recently closed a seed round of funding that included Khosla Ventures, Initialized Capital, and Charles River Ventures.

Voyage has been hot in Silicon Valley investor circles because of one big name linked to Udacity: Sebastian Thrun. Thrun, who founded the education startup, is also nicknamed the “Godfather of self-driving cars” for the work he did at Google and helped launch the self-driving car nanodegree program at Udacity.

Thrun, though, says he’ll have no connection with Voyage even though it’s spinning out of his company. “Because of personal conflicts, I have excused myself from any involvement in Voyage. I wish Oliver and his team all the best,” Thrun said in a statement to Business Insider.

The autonomous taxi startup wants to bring about the end goal where autonomous cars can carry people anywhere for a very low cost, Cameron said. It already has permission to deploy its self-driving cars to ferry passengers in a few places over the next few months, but Cameron declined to specify where.

“We want to deploy these not within five years, but very soon. We think in terms of weeks, not in terms of years or months,” he told Business Insider in an interview.

Pure guess: one reason Oliver Cameron decided to take this risk is because Udacity is open-sourcing it’s own self driving car project. All the code is there for all of us to use and improve. Including the new startup Voyage. And Oliver Cameron has a pretty good idea how successful the Udacity project is going to be.

Update: this week BMW has announced they plan to ship self driving cars in four years, in 2021. That’s similar to plans already announced by GM, Ford, Chrysler, Mercedes, Volvo and Chinese ride-sharing giant Didi Chuxing.

2017 Udacity Intersect: the future is closer than you think

We meat-bodies generally over-estimate short-term progress (one to three years) and underestimate medium-term technology progress (ten to twenty years). My particular interest is AI. That is partly because five decades ago Artificial Intelligence was my academic field at Carnegie Mellon. The logic-based AI that I was investigating with Herb Simon and Allen Newell is now known as GOFAI (good old-fashioned AI, that’s the AI that didn’t work very well). What motivates my current interest is that Machine Learning (ML), is starting to be really useful, and the rate of progress in narrow AI applications of ML is accelerating. You can see for yourself the rate of progress in ML in the explosion of speech recognition gadgets like Amazon Alexa, the voice service that powers the home Echo device. Now any of us can access voice and image recognition. In perhaps five years we will begin benefiting from Self Driving Cars (SDC) if we live in the right places.

I think that open-source, low-cost initiatives like Google’s release of TensorFlow mark a major inflection point in the rate of ML progress. A teenager can now prototype her ML-based idea to see if it really works. She doesn’t need to go to Sand Hill Road to raise venture capital. In fact, the VC community isn’t going to pay any attention to you unless you have already built your project to a level where it can be tested.

Another powerful indicator of the inflection point thesis is Udacity’s Nanodegree offerings. Outstanding example: the Self-Driving Car Engineer Nanodegree. For $800/term you can graduate with a qualification that dozens of leading companies are eager to hire. Here’s some of the hiring-partners of this SDC Engineer Nanodegree [they helped build the course]:


How can you learn more? Well, I suggest having a look at videos from the March 8th 2017 Udacity Intersect conference. This was a remarkable event, opening the window so all of us can get visibility on what is happening. The Computer History Museum was vibrating with the energy of companies recruiting Udacity students and Nanodegree graduates. Pretty much every panel and keynote of the Agenda was packed with tech industry insiders exchanging views about their projects, priorities and especially the people they want to hire.

I’m highlighting Udacity Intersect 2017 because the conference offers a concise and fun way to get a look into the future and behind the curtain. What are the leading tech companies thinking and doing? Where are we likely to be in 10 to 20 years?

You can find links to videos of every segment of the conference at the main Intersect page. If you’re not sure this is for you, please check out the final session Fireside Chat: Astro Teller and Sebastian Thrun. These 33 minutes gives you access to two of the leading innovators who have convinced me that “the future is closer than you think”.


Update: Udacity spins out self-driving taxi startup Voyage. Also, this week BMW has announced they plan to ship self driving cars in four years, in 2021. That’s similar to plans already announced by GM, Ford, Chrysler, Mercedes, Volvo and Didi.

Will competency replace traditional “seat time” qualifications?

“It’s scary for faculty,” Dr. Reilly says. “There’s a continuing sense that students can and do draw on so many sources of information that are now available at their fingertips. They don’t need to come to the monastery for four years and sit at the feet of the monks.”

“Now, I’m an old English professor who taught the Joyce course here at Madison two years ago,” he says. “The idea that you can’t understand Joyce unless you take it from Reilly three hours a week — that we faculty own the knowledge and anyone who’s going to be well educated has to get it from us — the world has changed so much that that’s no longer true.”

I’ve been following one of the leading innovators in Competency-Base education: Western Governors University. WGU was founded by a consortium of 19 states in 1997, offering a complete degree program based on assessments (not seat-time). Incoming students can have any accumulated credit hours evaluated for equivalency credit at WGU. At the time of writing of this NYT article there are now five US institutions offering Competency-Based degrees.

And it obsessively tracks metrics like this one: 95% of employers say WGU grads are as good as or better than those from anywhere else.

On average American teacher training programs are unspeakably bad. They serve as cash cows to subsidize the rest of the host school (US teachers must have a ‘ticket’ from one of these places). And they produce teachers who have had no training in the craft of teaching. But WGU appears to be an exception:

The best preparation program in the country for future high school teachers, according to new US News and World Report rankings, happens at an online university you probably haven’t heard of where students don’t take any classes.

The Western Governors University’s number-one spot surprised even the National Council on Teacher Quality, the advocacy group that worked with US News on the rankings. The high marks are not, however, a surprise to the nonprofit online university’s many fans — one of whom is President Obama.

Stanford’s Open Learning Initiative receives Gates Foundation grant

OLI Instructor Learning Dashboard [one of class level panels]

Thille explained that she considered three main resources through OLI: “What a student can do with their computer, what they can do with their peers, and what a student can do with an expert.” She also highlighted what she believes to be the key question in developing the program.

“What are the affordances and limitations of each of those resources and how can we blend them to create the best learning environment for that student at that point and time?” Thille said.

Very interesting — the Carnegie Mellon-developed Open Learning Initiative (OLI) has expanded to Stanford. Candace Thille, who founded OLI at CMU in 2002, has moved to Stanford’s Graduate School of Education. The Stanford project “is one of seven educational technology programs to split a $20 million fund from the Bill and Melinda Gates Foundation.” The Foundation has published an excellent 7 minute video that explains how OLI works for both instructors and students: Gates Foundation-The Open Learning Initiative.

Excerpts from the Stanford announcement:

OLI is an online educational platform that enhances classroom learning through digital modules that provide feedback to instructors. Students use the modules to learn and engage with material in an interactive online environment, while activities examine each student’s understanding of key concepts. The instructor uses the data from the platform to design the class and focus on important areas.

(…snip…) Originally, OLI focused mostly on assessing the cognitive process of learning. Since it has come to Stanford, the goal has been to build on its foundation while also exploring theoretical models and psychological assessments. Teams of disciplinary experts, learning researchers and software engineers build environments that can support classrooms anywhere that aim to provide the same experience as a top institution.

“The team designs these interactive environments, and the environments both support the learners but also collect the data to refine the learning environment,” Thille said.

The grant from the Bill and Melinda Gates Foundation will specifically go towards a module for an introductory statistics course, chosen for its high-enrollment and high-variability. Thille and her colleagues are building upon an existing statistics course and are integrating OLI into another online platform called edX. They will also be enhancing the environments with social-psychological interventions. Ultimately, the module will be used in thousands of classrooms across the nation.

Partners at other institutions offer faculty expertise for content development and use trial modules to gather data and improve strategies. These institutions include universities and community colleges in Maryland, North Carolina and California. The platform is designed to assess students’ specific sub-skills pertaining to larger concepts that they learn in the class, and design teams pool data to focus on refining less effective aspects of the online courses.

OLI also emphasizes the openness of every education environment it develops: Anyone can access its content. The program aims to lead the way in integrating technology into education. The enhanced statistics course which is already being taught at Stanford, and more courses will soon follow.

When Will Liberal Arts Colleges Take Advantage of the Open Learning Initiative?


(…snip…) In a recent study by Inside Higher Ed and the Babson Survey Research Group, about half of liberal arts professors said blended learning was inferior to the purely face-to-face kind; only 11 percent said it was superior. At the same time, more than a third said blended education was of equivalent quality to face-to-face. So, technically, the majority of liberal arts professors thought the blended version was as good as or better than the kind they have been delivering for centuries.

The good news is that 11% of liberal arts profs appreciate that advances like OLI can dramatically enhance learning. The not-so-good news is the 89% who think the status quo is OK. However, the better news is that the focus on learning is sharpening. Example: the encouraging news that Stanford has joined with CMU to advance both OLI content and technology.

So far I’m not having much success identifying the US liberal arts colleges that are making good use of the OLI work or similar learning technologies. Though the article is two years out of date, this Inside Higher Ed survey looks at what a couple of high visibility colleges are doing.

Online learning is no longer foreign to traditional universities, where courses formerly held in large lecture halls are migrating to the Web. But at residential liberal arts colleges, whose appeal often lies in the promise of small classes and regular face time with professors, online education has had a harder time gaining a foothold.

That could soon change. Several top-rated liberal arts colleges have begun experimenting with online course modules. Professors at those colleges hope the technology, which tutors students in certain concepts via artificially intelligent tutoring software in lieu of static textbooks or human lecturers, will help level the playing field for academically underprepared students while giving instructors more flexibility in planning their syllabuses.


“It is not our model — we’re more into human interaction,” says Cassidy, the Bryn Mawr provost. “But the data were persuasive.”

The “data” include multiple studies by researchers at Carnegie Mellon suggesting that when students are taught an introductory statistics using OLI course, in concert with human instruction, they learn the material as well or better than in a normal lecture course — but in half the time. More recently, researchers at the nonprofit Ithaka S+R found that a socioeconomically diverse sample of students at six public universities performed as well via a “blended” model of statistics education that replaced some seat time with independent work on the OLI platform.

The goal at Bryn Mawr is not to shorten the semester, as might be the case at a community college looking to speed the rotation of the academic calendar for revenue purposes, or to satisfy a completion-agenda push to increase graduation rates. And the college has no plans to build a large online operation on top of its bricks-and-mortar campus. “Fundamentally, we’re secure with our model” of small-scale residential education, says Cassidy.

But Bryn Mawr’s foray into using online course modules hints at the online learning tools that might have a place in the liberal arts. Joey King, the president of the National Institute for Technology in Liberal Education, had not heard of any cases where liberal arts professors were integrating OLI modules or similar technology into their syllabuses. But he said that Bryn Mawr’s leadership and pledge to support similar projects at peer colleges might catalyze the spread of new, “blended” approaches in that sector.

“In this community it’s kind of the norm that somebody will take the lead and just do it,” says King, “and, once it’s easy, people find it more compelling as far as picking it up at other institutions.”

Faculty at liberal arts colleges might be more amenable to accepting help from robo-tutors than instructors at other types of institutions, King added. Such technology often spurs anxieties among instructors that the institution is gradually supplanting them with teaching machines. But the liberal arts colleges’ value proposition still hinges on the presence of human professors. According to Spohrer, 100 percent of Bryn Mawr faculty members who used the technology said they plan to use it again.

Please let me know if you have references/studies of liberal arts colleges that have made significant progress modernizing their learning processes.


New York charter school gets results from teacher incentives, pay, supervision

This Gates Foundation funded study is encouraging. 

(…snip…) The typical teacher in New York with five years’ experience makes between $64,000 and $76,000. The charter school, known as TEP, would pay much more [$125,000 Ed]. But in exchange, teachers, who are not unionized, would accept additional responsibilities, and the school would keep a close eye on their work.

Four years later, students at TEP score better on state tests than similar students elsewhere. The differences were particularly pronounced in math, according to a new study from Mathematica Policy Research. (The study was funded by the Gates Foundation.) After four years at the school, students had learned as much math as they would have in 5.6 years elsewhere:



The gains erased 78 percent of the achievement gap between Hispanic students and whites in the eighth grade.


The results are important in part because TEP also appears to have sidestepped some common concerns about charter schools. They didn’t expel or suspend students out of school in the first four years. There is no evidence that the school encouraged problem students to leave or transfer on their own. And the students who attended were roughly as likely to be low-income, and to have had similar levels of academic achievement before they arrived. They could still differ in other ways — they could have more involved parents, who get them into the charter school lottery, for example — but TEP doesn’t present some of the obvious factors that help explain other charter schools’ success.

More details…

Parents: “Don’t send your kid to the Ivy League”


Is there anything that I can do, a lot of young people have written to ask me, to avoid becoming an out-of-touch, entitled little shit? I don’t have a satisfying answer, short of telling them to transfer to a public university. 

If you have a child that is likely to attend a US four-year college, read on. William Deresiewicz is the author of Excellent Sheep: The Miseducation of the American Elite and The Way to a Meaningful Life. His recent essay at New Republic is a must-read for both parents and prospective students. Deresiewicz finished a ten-year stint teaching at Yale in 2008, where he begins by describing admissions:

In the spring of 2008, I did a daylong stint on the Yale admissions committee. We—that is, three admissions staff, a member of the college dean’s office, and me, the faculty representative—were going through submissions from eastern Pennsylvania. The applicants had been assigned a score from one to four, calculated from a string of figures and codes—SATs, GPA, class rank, numerical scores to which the letters of recommendation had been converted, special notations for legacies and diversity cases. The ones had already been admitted, and the threes and fours could get in only under special conditions—if they were a nationally ranked athlete, for instance, or a “DevA,” (an applicant in the highest category of “development” cases, which means a child of very rich donors). Our task for the day was to adjudicate among the twos. Huge bowls of junk food were stationed at the side of the room to keep our energy up.

The junior officer in charge, a young man who looked to be about 30, presented each case, rat-a-tat-tat, in a blizzard of admissions jargon that I had to pick up on the fly. “Good rig”: the transcript exhibits a good degree of academic rigor. “Ed level 1”: parents have an educational level no higher than high school, indicating a genuine hardship case. “MUSD”: a musician in the highest category of promise. Kids who had five or six items on their list of extracurriculars—the “brag”—were already in trouble, because that wasn’t nearly enough. We listened, asked questions, dove into a letter or two, then voted up or down.

With so many accomplished applicants to choose from, we were looking for kids with something special, “PQs”—personal qualities—that were often revealed by the letters or essays. Kids who only had the numbers and the résumé were usually rejected: “no spark,” “not a team-builder,” “this is pretty much in the middle of the fairway for us.” One young person, who had piled up a truly insane quantity of extracurriculars and who submitted nine letters of recommendation, was felt to be “too intense.” On the other hand, the numbers and the résumé were clearly indispensable. I’d been told that successful applicants could either be “well-rounded” or “pointy”—outstanding in one particular way—but if they were pointy, they had to be really pointy: a musician whose audition tape had impressed the music department, a scientist who had won a national award.

“Super People,” the writer James Atlas has called them—the stereotypical ultra-high-achieving elite college students of today. A double major, a sport, a musical instrument, a couple of foreign languages, service work in distant corners of the globe, a few hobbies thrown in for good measure: They have mastered them all, and with a serene self-assurance that leaves adults and peers alike in awe. A friend who teaches at a top university once asked her class to memorize 30 lines of the eighteenth-century poet Alexander Pope. Nearly every single kid got every single line correct. It was a thing of wonder, she said, like watching thoroughbreds circle a track.

These enviable youngsters appear to be the winners in the race we have made of childhood. But the reality is very different, as I have witnessed in many of my own students and heard from the hundreds of young people whom I have spoken with on campuses or who have written to me over the last few years. Our system of elite education manufactures young people who are smart and talented and driven, yes, but also anxious, timid, and lost, with little intellectual curiosity and a stunted sense of purpose: trapped in a bubble of privilege, heading meekly in the same direction, great at what they’re doing but with no idea why they’re doing it.

My personal perspective is that an engineering or science degree from Stanford, Carnegie Melon or similar schools is a very good thing for the students that can truly afford the cost. But I also think an elite liberal arts education is a questionable value proposition – where the “product” the parents are buying isn’t necessarily what you think it is. Read the Deresiewicz essay.

Joel Cohen: Malthus Miffed: Are People the Problem, the Solution, or Both?

I highly recommend that you inspect Floating University’s Great Big Ideas: An Entire Undergraduate Education While Standing on One Foot. 

In the Fall of 2011 Big Think teamed up with the Jack Parker Corporation to launch The Floating University, an online educational initiative that debuted at Harvard University, Yale University, and Bard College. Seeking to upset the status quo, evolve the structure of higher education, and democratize access to the world’s best thinkers, FU’s inaugural course, Great Big Ideas, became the most requested class at all three schools where it was offered.(…snip…)

There are twelve lectures, each taught by a leader in the field who is also a great teacher. The first lecture of the series is the captioned Malthus Miffed by Joel Cohen, a mathematical biologist and a professor of populations. It is a suitable topic for the first lecture because an understanding of demography is one of the foundations for understanding how the world works, and especially what policies are likely to succeed (e.g., immigration, development, climate).

Prof. Cohen really is a great teacher – a skill achieved by investing a lot of energy in developing the craft, including practice. Even if you don’t think you are interested in demographics I predict you will be glued to your screen for the duration of this lecture. The course package includes Readings and Discussion Questions. 


Students hope to gain access to competent teachers

Students Matter has serious money for a serious legal fight to break the union stranglehold. If they win the California appeal it will become possible to fire terrible teachers

LOS ANGELES — They have tried and failed to loosen tenure rules for teachers in contract talks and state legislatures. So now, a group of rising stars in the movement to overhaul education employment has gone to court.

In a small, wood-paneled courtroom here this week, nine public school students are challenging California’s ironclad tenure system, arguing that their right to a good education is violated by job protections that make it too difficult to fire bad instructors. But behind the students stand a Silicon Valley technology magnate who is financing the case and an all-star cast of lawyers that includes Theodore B. Olson, the former solicitor general of the United States, who recently won the Supreme Court case that effectively overturned the state’s ban on same-sex marriage.

“Children have the right to access good education and an effective teacher regardless of their circumstances,” said David F. Welch, the telecommunications entrepreneur who spent millions of his own dollars to create Students Matter, the organization behind the lawsuit. The group describes itself as a national nonprofit dedicated to sponsoring litigation of this type, and the outcome in California will provide the first indication of whether it can succeed.

John E. Deasy, the schools superintendent of Los Angeles. Monica Almeida/The New York Times

At issue is a set of rules that grant permanent employment status to California teachers after 18 months on the job, require a lengthy procedure to dismiss a teacher, and set up a seniority system in which the teachers most recently hired must be the first to lose their jobs when layoffs occur, as they have regularly in recent years.

Teachers’ unions, which hold powerful sway among lawmakers here, contend that the protections are necessary to ensure that teachers are not fired unfairly. Without these safeguards, the unions say, the profession will not attract new teachers.

“Tenure is an amenity, just like salary and vacation, that allows districts to recruit and retain teachers despite harder working conditions, pay that hasn’t kept pace and larger class sizes,” James M. Finberg, a lawyer for the California teachers’ unions, said this week in his opening statement in court.

The monthlong trial promises to be a closely watched national test case on employment laws for teachers, one of the most contentious debates in education. Many school superintendents and advocates across the country call such laws detrimental and anachronistic, and have pressed for the past decade for changes, with mixed success. Tenure for teachers has been eliminated in three states and in Washington, D.C., and a handful of states prohibit seniority as a factor in teacher layoffs. But in many large states with urban school districts, including California and New York, efforts to push through such changes in the legislature have repeatedly failed.

While several lawsuits demanding more money for schools have succeeded across the country, the California case is the most sweeping legal challenge claiming that students are hurt by employment laws for teachers. The case also relies on a civil rights argument that so far is untested: that poor and minority students are denied equal access to education because they are more likely to have “grossly ineffective” teachers.

Judge Rolf Michael Treu, of Los Angeles County Superior Court, will decide the nonjury trial. His ruling will almost certainly be appealed to the State Supreme Court.


KNOWOSPHERE: can MOOCS make a difference?

I think Andy Revkin's KNOWOSPHERE is a useful framing of one of the core development challenges. 18 months ago Andy wrote about how even South African students couldn't access higher education. SA is relatively rich compared to many neighbors.

I also think Tyler Cowen's “Average is Over” is fundamentally correct. So how are the Bottom Billion going to find jobs that lead out of the bottom? The only scalable, affordable pathway I've been able to think of are MOOCS. Remember that the top 72 students in the first Stanford online AI class were NOT Stanford students!

Here's Andy from 2012: What Can U.S. Universities Do About a Student Stampede in Johannesburg?

…To me, there is nothing more tragic than seeing young people who are already eager to learn denied that chance — whether through inequity created by poverty or simply, as in this case, the lack of infrastructure. (I had that same feeling when I first saw photos of kids, lacking electricity in their slum dwellings, doing homework under the lights in an airport parking lot in Guinea.)

From South Asia through much of the Middle East and sub-Saharan Africa, it’d be impossible to build schools or train teachers fast enough to keep up with the “youth bulge” that has given humanity more than a billion teenagers either to nurture or tame — the difference depending largely on access to education beyond elementary grades.

But in these same places, explosive expansion in mobile phone subscriptions and fast-dropping costs for smart phones provide the architecture for a partial end run around such bottlenecks. That’s why the decision by the Massachusetts Institute of Technology to open more courses to online users is probably just a taste of what’s to come. [Stanford University has had remarkable outcomes, as well.]

What’s needed now is the educational equivalent to Paul Polak’s work fostering progress in rural agrarian communities in poor places. His mantra is “design for the other 90 percent.”

Universities in the developed world seeking a place (and a business model) in a century in which knowledge is no longer cached in ivory towers would do well to find ways to “educate for the other 90 percent.”

So here's the question: what do we have to do to enable Nigeria to Somalia to leverage all those free MOOCS into useful education and brainwork jobs? Just a smartphone is not sufficient. We are starting to see some enabling models in the rich world — WGU Western Governors University is great example. Check it out, how could it be adapted to South Africa?