ZeV > Recent Status Updates

Multi-armed bandit problems are everywhere. How well do people solve these problems? In general, it seems that people tent to over explore (i.e., under exploit as a result).

Clinical trial and the selection of treatments as a multi-armed bandit problem, trying to get the better treatments to people even while an experiment is in progress. (History of development of such methods.)

Algorithms that offer the guarantee of minimal regret ... of the ones they've discovered, the most popular are known as Upper Confidence Bound algorithms.

Jeff Bezos's "regret minimization framework": So I wanted to project myself forward to age 80 and say, "Okay, now I'm looking back on my life. I want to have minimized the number of regrets I have. ... I knew the one thing I might regret is not ever having tried."

The Gittins index provides a formal, rigorous justification for preferring the unknown, provided we have some opportunity to exploit the results of what we learn from exploring ... the unknown has a chance of being better, even if we actually expect it to be no different, or if it's just as likely to be worse. The untested rookie is worth more than the veteran of seemingly equal ability, since we know less about him.

Exploration vs. exploitation: A sobering property of trying new things is that the value of exploration, of finding a new favorite, can only go down over time, as the remaining opportunities to savor it dwindle. Discovering an enchanting cafe on your last night in town doesn't give you the opportunity to return. ... The flip side is that the value of exploitation can only go *up* over time.

Intuitively, we think that rational decision-making means exhaustively enumerating our options, weighing each one carefully, and then selecting the best. But in practice, when the clock--or the ticker--is ticking, few aspects of decision-making (or thinking more generally) are as important as one: when to stop.

The number of robberies you should carry out before stopping is roughly equal to the chance you get away, divided by the chance you get caught.

Following the optimal strategy, it is also that there is 37% chance of hiring the best applicant; it's one of the problem's curious mathematical symmetries.

The look-then-leap rule: You set a predetermined amount of time for "looking"-- that is, exploring your options, gathering data--in which you categorically don't choose anyone, no matter how impressive. After that point, you enter the "leap" phase, prepared to instantly commit to anyone who outshines the best applicant you saw in the look phase.

The 37% rule of the optimal stopping problem.

On Mishima. Alberto Moravia. Norman Mailer. Marguerite Yourcenar.

What should you read next?

Book recommendations: books@egoistheenemy.com
or www.EgoIsTheEnemy.com/books.

Monthly book recommendations: ryanholiday@gmail.com with "Reading List E-mail" in the subject line or sign up at ryanholiday.net/reading-newsletter

"There's a quote from Bismarck that says, in effect, any fool can learn from experience. The trick is to learn from *other people*'s experience."

"training was like sweeping the floor. Just because we've done it once, doesn't mean the floor is clean forever. Every day the dust comes back. Every day we must sweep." -- Daniele Bolelli

"I don't like work--no man does--but I like what is in the work--the chance to find yourself." -- Joseph Conrad

"In failure or adversity, it's so easy to hate. Hate defers blame. It makes someone else responsible. It's a distraction too; we don't do much else when we're busy getting revenge or investigating the wrongs that have supposedly been done to us ... Does this get us any close to where we want to be? No. It just keeps us where we are--or worse, arrests our development entirely."

"This is characteristics of how great people think. It's not that they find failure in every success. They just hold themselves to a standard that exceeds what society might consider to be objective success. Because of that, they don't much care what other people think; they care whether they meet their own standards. And these standards are much, much higher than everyone else's."

"I never look back, except to find out about mistakes ... I only see danger in thinking back about things you are proud of." -- Elisabeth Noelle-Neumann

"He who fears death will never do anything worthy of a living man" -- Seneca "Alter that: He who will do anything to avoid failure will almost certainly do something *worthy of a failure* ... The only real failure is abandoning your principles. Killing what you love because you can't bear to part from it is selfish and stupid. If your reputation can't absorb a few blows, it wasn't worth anything in the first place."

"Most trouble is temporary ... unless you make that not so. Recovery is not grand, it's one step in front of the other. Unless your cure is more of the disease."

"The problem is that when we get our identity tied up in our work, we worry that any kind of failure will then say something bad about us *as a person*. It's a fear of taking responsibility, of admitting that we might have messed up. It's the sunk cost fallacy. And so we throw good money and good life after bad and end up making everything so much worse."

"In the end, the only way you can appreciate your progress is to stand on the edge of the hole you dug for yourself, look down inside it, and smile fondly at the bloody claw prints that marked your journey up the walls."

(con't) "Ego often causes the crash and then blocks us from improving."

"A look at history finds that these events seem to be defined by three traits: 1. They almost always came at the hands of some outside force or person. 2. They often involved things we already knew about ourselves, but were too scared to admit. 3. From the ruin came the opportunity for great progress and improvement. Does everyone take advantage of that opportunity? Of course not." (con't)

"we cannot be humble except by enduring humiliations." -- The Reverend William A. Sutton