How spin becomes gospel

How and why do so many health myths transform into everyday wisdom? Perhaps because academic press releases are exaggerated or misrepresented when they're written up in news stories.

The goal of a press release around a scientific study is to draw attention from the media, and that attention is supposed to be good for the university, and for the scientists who did the work. Ideally the endpoint of that press release would be the simple spread of seeds of knowledge and wisdom; but it's about attention and prestige and, thereby, money. Major universities employ publicists who work full time to make scientific studies sound engaging and amazing. Those publicists email the press releases to people like me, asking me to cover the story because "my readers" will "love it." And I want to write about health research and help people experience "love" for things. I do!  

Across 668 news stories about health science, the Cardiff researchers compared the original academic papers to their news reports. They counted exaggeration and distortion as any instance of implying causation when there was only correlation, implying meaning to humans when the study was only in animals, or giving direct advice about health behavior that was not present in the study. They found evidence of exaggeration in 58 to 86 percent of stories when the press release contained similar exaggeration. When the press release was staid and made no such errors, the rates of exaggeration in the news stories dropped to between 10 and 18 percent.

Even the degree of exaggeration between press releases and news stories was broadly similar.

One golden rule: never believe a press release.

Of course, since most research papers I encounter online are behind a paywall, it's difficult to compare them to the news stories representing them. Why are so many research papers paywalled? I can't imagine the revenue to be more than a trifle so why not just let more eyeballs at it to amplify one's fame instead? Perhaps someone with more knowledge of that world can explain the economics.

The poker-playing AI

Researchers recently announced that they'd developed a poker bot that had weakly solved heads-up limit hold'em poker. That is, essentially the poker bot is unbeatable in a fair game, luck notwithstanding.

But although bluffing looks like a very human, psychological element of the game, it’s not. You can calculate how to bluff optimally. “Bluffing falls out of the mathematics of the game”, says Bowling. If you’re dealt a jack, say, it is possible to figure out how often you should ideally bluff with it. Some of the early pioneers of game theory, such as John von Neumann, aimed to develop mathematical strategies for bluffing. 

The real challenge for a poker algorithm is dealing with the immense number of possible ways the game can be played. Bowling and colleagues have looked at one of the most popular forms, called Texas Hold’em, in which the dealer deals two cards to each player (face down) and also a set of face-up “community cards”. Players can bet, raise or fold after each deal. With just two players, the game becomes Heads-up, and it is a “limit” game when it has fixed bet sizes and a fixed number of raises. There are 3.16×10^17 states that HULHE can reach, and 3.19×10^14 possible points where a player must make a decision. 

The new algorithm involves calculating all possible decisions in advance, so that they can just be looked up as a game proceeds. This is done in a learning process: the strategy begins by making decisions randomly, and is then updated by experience as the algorithm attaches a “regret” value to each decision depending on how poorly it fared. It takes a little more than 1500 training rounds to make the program essentially invincible.

What caught my eye is how they solved the problem. One of the keys to dealing with the problem was actually just a data-compression solution.

The other crucial innovation was the handling of the vast amounts of information needing to be stored to develop and use the strategy – of the order of 262 terabytes. This volume of data demands disk storage, which is slow to access. The researchers figured out a data-compression method that reduces the volume to a more manageable 11 TB, and which adds only 5% to the computation time from the use of disk storage.

You can try playing this new poker bot online.

It's an impressive advance, but I wonder how long until they can solve multi-player no-limit Texas Hold'em. It's a much more complex game than heads-up limit hold'em.

Rate-limiting steps

Bill and Melinda Gates are doing it right. From 1990 to recently, the childhood mortality rate has been cut in half. Why? Cause they pay attention to the things that matter. What kills kids worldwide? Malaria - and they’re hard at work on a vaccine. No worries about phantom illnesses or the craze of the week. Instead, they’ve got a solid focus on the things that actually matter.

So what do they propose to keep things going? Again – rate limiting steps...

How Bill and Melinda Gates are making the most of their philanthropy. It's no coincidence that one of GiveWell's top-rated charities is the Against Malaria Foundation.

Rate-limiting steps is a concept with metaphoric value in lots of spheres in life.

Buzzfeed: giving you what you'll share

The heavyweight personalization that social networks do either through something like the follow graph on Twitter or through more algorithmic approach on Facebook is designed to give people the stuff that they want. What we're focused on is giving people stuff that they think is worth sharing with other people. We want the stories we're doing to have the biggest possible impact. So if we do personalization, it would be more of a personalization about what you’re most likely to share or discuss with your friends.

That's Jonah Peretti on what type of personalization Buzzfeed focuses on, emphasis mine on what is a subtle but important focusing distinction for the service.

In the traditional news bundle, say in old school print newspapers, the mix of serious versus entertaining content was weighted much more to the former. Now that technology has allowed the creation of more personalized bundles of information, sites like Buzzfeed and the social networks like Facebook and Twitter are showing us people's natural preference in the mix of heavy versus light, and it turns out that ratio is much more weighted towards the fun.

As in the newspaper days, the entertaining content still subsidizes, to a large extent, the serious journalism. Buzzfeed is starting to do some original reporting, but more likely than not it's ad revenue from listicles and the more “frivolous” content that will pick up the tab for both. Plus ça change.

How Los Angeles came to have the best Chinese Food in America

Having lived in most of the major U.S. cities and sampled their best Chinese food, being Chinese-American, loving Chinese food, having a mom who taught Chinese cooking, having eaten Chinese food in China and Taiwan and Hong Kong, I throw my lot in with this thesis: the best Chinese food in America is in the Eastern suburbs of Los Angeles.

New Yorkers think they know the real thing when it comes to Chinese food. It has been a topic of hot debate. A lot of folks like to cite Flushing, where there are some legitimate regional specialists. But when it comes to quality, it is Los Angeles that reigns supreme—yes, better than Flushing and Vancouver.

“For probably 140 years, the best Chinese food in the U.S. was in San Francisco,” David R. Chan, a Los Angeles attorney and Chinese food hobbyist says. Chan has eaten at more than 6,500 Chinese restaurants since 1951 and has been documenting his progress on a massive spreadsheet, recording the date and address of his visits. Chan’s interest lies in systematics. A third-generation Taishanese-American and one of the first students enrolled at UCLA’s Asian-American program, Chan uses his spreadsheet as a lens to observe the progression of the Chinese diaspora in America. Food after all, is at the apex of Chinese culture.

It wasn’t until the late 1980s that the Bay Area lost its crown, and all the action shifted towards the San Gabriel Valley. “That’s when Chinese food in Los Angeles experienced a major upswing,” says Chan.

If New York is home to the largest population of Chinese-Americans in the States, why, then, does Los Angeles still hold the mantle for best Chinese food? Chef pedigree, regional diversity, and a strong local food community are part of the story.

As I've written before, I think restaurant quality today is largely a supply-side problem, and that applies even more so with an ethnic cuisine like Chinese food in America. For a variety of reasons, if you're a great Chinese chef, living in suburbs like Arcadia, San Gabriel, and Monterey Park is highly desirable.