Sunday, July 24, 2016

Scifoo 2016

Photos from Palo Alto and Scifoo 2016. We weren't allowed to take photos inside the Googleplex.

Wednesday, July 20, 2016

Farewell Asia, Hello Scifoo

Apologies for the lack of blog posts. I've been on the road in Asia and quite busy for the past week. I head back to the bay area for Scifoo this weekend. See you there!

The tipping point

This is only the beginning. To serious people who read this blog, but may have been confused over the past 5+ years about things like missing heritability, genomic prediction, complex genetic architecture, gloomy prospects: isn't it about time to consider updating your priors? Read all about it here. Genetic scoring predicts how children do at school

Professor Robert Plomin, senior author, called the study “a tipping point for predicting individuals’ educational strengths and weaknesses from their DNA”. It is published in the journal Molecular Psychiatry.

An individual’s “polygenic score” is based on the presence or absence of 20,000 common DNA variants across many different genes. Each has a tiny effect on its own but together they explain 10 per cent of the variation in children’s educational attainment at the age of 16.
It's important to note that 9 percent of variance accounted for implies that the resulting polygenic score predictor would correlate about 0.3 with actual educational achievement. This is similar to the correlation between high school and college grades at a typical flagship state university (note there is some restriction of range) -- not a weak predictor by the standards of social science.
Predicting educational achievement from DNA

Nature Molecular Psychiatry 19 July 2016; doi: 10.1038/mp.2016.107

A genome-wide polygenic score (GPS), derived from a 2013 genome-wide association study (N=127,000), explained 2% of the variance in total years of education (EduYears). In a follow-up study (N=329,000), a new EduYears GPS explains up to 4%. Here, we tested the association between this latest EduYears GPS and educational achievement scores at ages 7, 12 and 16 in an independent sample of 5825 UK individuals. We found that EduYears GPS explained greater amounts of variance in educational achievement over time, up to 9% at age 16, accounting for 15% of the heritable variance. This is the strongest GPS prediction to date for quantitative behavioral traits. Individuals in the highest and lowest GPS septiles differed by a whole school grade at age 16. Furthermore, EduYears GPS was associated with general cognitive ability (~3.5%) and family socioeconomic status (~7%). There was no evidence of an interaction between EduYears GPS and family socioeconomic status on educational achievement or on general cognitive ability. These results are a harbinger of future widespread use of GPS to predict genetic risk and resilience in the social and behavioral sciences.

Saturday, July 16, 2016

Monday, July 11, 2016

My neighbor's kid is a designer baby (or not)

I happened to see Knoepfler's book at the HKUST bookstore yesterday. He's a stem cell researcher at UC Davis.
GMO Sapiens: the Life-Changing Science of Designer Babies is an exciting new book which takes a look at the cutting edge biotechnology trends that are making human genetic modification and cloning a more practical potential reality in the coming years. The convergence of advances in bioengineering, genetics, genomics, assisted reproduction and stem cells have brought us to the astonishing point where literally making new forms of human beings is possible. Surprisingly, some people and organizations are strong advocates for allowing genetic modification of humans while others feel equally passionately opposed. Meanwhile, the general public is largely unaware that we have reached this remarkable turning point in human history when we can literally change what it means to be human.

As much as the public is engaged in the topic of genetically modified organism (GMO) foods, this book will capture their imaginations and fire their passions by educating them on the potential coming of GMO people while, at the same time, educating them.

Sunday, July 10, 2016


By coincidence, they're hosting the International Math Olympiad here. There are teams of kids from all over the world, but less geeky looking than I expected.

This deck of cards was created by the HKUST math chair, formerly of MSU.

Tuesday, July 05, 2016

OUR FINAL INVENTION: Artificial Intelligence and the End of the Human Era (Future Strategist podcast)


My long fascination with Artificial Intelligence came to a head in 2000, when I interviewed inventor Ray Kurzweil, roboticist Rodney Brooks, and sci-fi legend Arthur C. Clarke. Kurzweil and Brooks were casually optimistic about a future they considered inevitable – a time when we will share the planet with intelligent machines. “It won’t be some alien invasion of robots coming over the hill,” Kurzweil told me, “because they’ll be made by us.” In his compound in Sri Lanka, Clarke wasn’t so sure. “I think it’s just a matter of time before machines dominate mankind,” he said. “Intelligence will win out.”

Intelligence, not charm or beauty, is the special power that enables humans to dominate Earth. Now, propelled by a powerful economic wind, scientists are developing intelligent machines. Each year intelligence grows closer to shuffling off its biological coil and taking on an infinitely faster and more powerful synthetic one. But before machine intelligence matches our own, we have a chance. We must develop a science for understanding and coexisting with smart, even superintelligent machines. If we fail, we’ll be stuck in an unwinnable dilemma. We’ll have to rely on the kindness of machines to survive. Will machines naturally love us and protect us?

Should we bet our existence on it?

Our Final Invention is about what can go wrong with the development and application of advanced AI. It’s about AI’s catastrophic downside, one you’ll never hear about from Google, Apple, IBM, and DARPA. I think it’s the most important conversation of our time, and I hope you’ll join in.

OUR FINAL INVENTION: Artificial Intelligence and the End of the Human Era

Artificial Intelligence helps choose what books you buy, what movies you see, and even who you date. It’s in your smart phone, your car, and it has the run of your house. It makes most of the trades on Wall Street, and controls our transportation, energy, and water infrastructure. Artificial Intelligence is for the 21st century what electricity was for the 20th and steam power for the 19th.

But there’s one critical difference — electricity and steam will never outthink you.

The Hollywood cliché that artificial intelligence will take over the world could soon become scientific reality as AI matches then surpasses human intelligence. Each year AI’s cognitive speed and power doubles — ours does not. Corporations and government agencies are pouring billions into achieving AI’s Holy Grail — human-level intelligence. Scientists argue that AI that advanced will have survival drives much like our own. Can we share the planet with it and survive?

Our Final Invention explores how the pursuit of Artificial Intelligence challenges our existence with machines that won’t love us or hate us, but whose indifference could spell our doom. Until now, intelligence has been constrained by the physical limits of its human hosts. What will happen when the brakes come off the most powerful force in the universe?

See also my interview on this podcast, a project of economics professor James Miller. The two key technologies which will shape the future are AI and genetic engineering.

Saturday, July 02, 2016

Chaos Monkeys: physics to Goldman to Y Combinator to Twitter to Facebook

Highly recommended! I blogged about this guy 5 years ago here: From physics to Goldman to Y Combinator. The book is hilarious and pretty accurate, AFAICT. I don't know much about Facebook corporate culture or that particular era of ad monetization, but the finance and startup stuff all rings true.
The reality is, Silicon Valley capitalism is very simple:
Investors are people with more money than time.
Employees are people with more time than money.
Entrepreneurs are the seductive go-between.
Startups are business experiments performed with other people's money.

I was a Berkeley PhD student in physics when the first dot-com bubble grew to bursting and popped around 2001. Between the month-long backpacking trips and the telenovela-esque romances, I switched thesis topic three times, and felt my twenty-something vitality slipping away in academic wankery. Inspired by Michael Lewis’ Liar’s Poker and the prior example of many a failed physicist, I looked for a Wall Street gig as a way out. Very improbably, I landed a job on the trading desk of Goldman Sachs, earning twice what my tenured professor made, pricing and modeling credit derivatives at ground-zero of the credit bubble. I may have owned one pair of lace-up shoes at the time, but I got used to speaking in quantities of hundreds of millions of dollars, and thinking a million was a ‘buck’, i.e., a rounding error for most purposes. I was very far away indeed from Berkeley.

Right around 2008, when Lehman Brothers and Bear Stearns blew up, I knew the financial jig would be up for a while (and possibly forever), unlike most of my colleagues, who seemed to think orgies of rapacious greed lasted forever. The only piece of the US economy that would be spared the apocalypse was clear in my mind: the Bay Area tech of my languid grad school days, and all that VC money that (hopefully) hadn’t touched the mortgage bubble..

Two weeks later, I started as employee number seventy-something at a venture-backed advertising startup so incompetent and vile I’ll save the historical distaste for later. Bookended as it was by experiences at Facebook and Goldman, my time there was instructive in its awfulness and how not to run a company. But there was one piece of upside: I learned how online advertising worked, specifically its ad exchange variants. As a ‘research scientist’ I tortured every piece of data until it confessed, and used it to predict user behavior, value of media purchased, and optimal bids in the largest ad auctions in the world. Dull stuff you might say, but it’s what pays for the Internet, and it would set me light-years ahead of anyone inside Facebook Ads, when the time came.

But we’re jumping ahead.

Along with the two best engineers at Shitty Unnamed Company, I applied and was accepted to Y Combinator, the Valley’s leading startup incubator. We pitched some wild, ridiculous idea around local businesses that was doomed from the start, which eventually morphed into a novel tool for managing Google search campaigns for small businesses. The tool was beautiful, innovative, and didn’t make us a dime. More bad news: We got vindictively and frivolously sued by Shitty Company and fought an existential legal battle we narrowly won by being lying, ruthless little shits. We couldn’t raise money. We had co- founder and morale issues. Every ill that plagues early-stage startups visited us in turn, like some admonitory biblical tale about what happens if you fuck with the Israelites. ...

... Every startup entrepreneur faces the immense disadvantage of playing a crooked, complex game for the first time, against a world composed mostly of masters. Arrayed against you is an army of wily, self-interested venture capitalists who know term sheets better than their wife’s ass. Or seductive sales execs who could make pedophilia and genocide enforceable via a legal contract. Or petulant co-founders with hidden agendas and momentarily suppressed grievances. Or ungrateful employees who are exploiting your startup until they can start their own. Or thick-headed journalists with urgent deadlines who just want you for a misleading quote. You get trounced again and again, and the only hope is that you learn something of the game before expiring. This is your principal challenge as a first-time entrepreneur: to learn the game faster than you burn cash and relationships.

Blog Archive