Regret minimization as life strategy

Regret minimization (technically regret minmax – minimize maximal regret) is a useful mental model for decision making.

Recently Tyler Cowen discussed it in an excellent interview The Complacent Class, Sex Robots, and Deathbed Regrets: A Conversation with Tyler Cowen (skip to 49:45, but the whole thing is recommended).

There is no transcript so here are my loose citation notes:

If you lived an “optimal” life there would be a lot of regret at the end of it.

Regret minimization is not the best life strategy.

Your deathbed perspective is not the best/most relevant metric of anything:

you perception is at lowest
your cognition is at lowest
your memory is the worst
you are not responsible for anything

Compare with Jeff Bezos’ decision strategy:

“The framework I found, which made the decision incredibly easy, was what I called — which only a nerd would call — a “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 that when I was 80 I was not going to regret having tried this. I was not going to regret trying to participate in this thing called the Internet that I thought was going to be a really big deal. I knew that if I failed I wouldn’t regret that, but I knew the one thing I might regret is not ever having tried. I knew that that would haunt me every day, and so, when I thought about it that way it was an incredibly easy decision.”

The same model is cast in different light. I think Cowen and Bezos would actually mostly agree – “local” regret minimization is premature optimization.

  Cowen Bezos
Apparently says that regret min strategy is… bad good
which regret he means experiential self: real-time regret remembered self – regret at life end
does the utility of
death-bed-me matter
not much yes

The only thing, that is not quite reconciled is – how much should you take into account your “death-bed” personas preferences?

Which reminds me of a famous quote:

Twenty years from now you will be more disappointed by the things you didn’t do than by the ones you did do

— apparently NOT Mark Twain

Reading and publishing

One of the recurring topics of this blog is AI and automation. So it is only natural that I also automatized my blogging: I introduce you to my new blog:

It will collect various snippets that I’ve found interesting during my reading.

The blog is stochastic – every day it randomly decides if there will be a new post or not and then it randomly picks the snippet.

One aspect that I like about it is that it seamlessly integrates the act of reading with the act of publishing.

The other aspect that I like is that it might be actually good, unlike this blog 🙂

In any case I’ll subscribe to it – it will be nice to receive this random reminders from my past self.

Three Types of Symmetry

Past 6 months brought a lot of travel, relocation to a new city & country, new job…  now let’s try to get back into good habits like writing.

From 101 Things I Learned in Architecture School I finally understood the appeal of asymmetrical architecture and why is it so beloved by modern architects. This appreciation resonated with a 900 year old echo when we visited Angkor Wat.

1. Static symmetry

The most basic type of symmetry is the one we all imagine when one mentions “symmetry”. Invariance of shapes (or systems) under certain transformations (rotation, reflection, scaling) is deeply appealing to us. Like the layout of the Angkor Wat complex.


2. Symmetry, broken

The breaking of symmetry is however a quintessential physical mechanism that gives rise, among others, to replicating complex patterns balancing on the edge of chaos (aka “life”).

The breaking of symmetry between past and future gives rise to the arrow of time. And where is time there are stories – such as the The Battle of Kurukshetra. Static symmetry of the army formations is ultimately broken to be able to tell the story of the big battle.


3. Dynamic balance – the symmetry of asymetry

Finally, we arrive at the least obvious form of symmetry – the dynamical balance.

From 101 Things I Learned in Architecture School:

Balance is inherent in a symmetrical composition, but asymmetrical com-
positions can be either balanced or unbalanced. Consequently, asymmetry tends to
require a more complex and sophisticated understanding of wholeness.

An example – a dancer is in dynamic balance, yet the posture is not symmetrical in the static sense.


A spinning dancer is in dynamical balance. To keep with the Angkor Wat theme I could have showed a picture of dancing Apsaras. I found this optical illusion however extra interesting, because in addition to dynamical balance it demonstrates a dynamical clockwise/anti-clockwise symmetry in rotation.

This concept too hasn’t been unknown to the builders of Angkor Wat in the 11th century. My favourite bas relief is The Churning of Ocean of Milk.


It shows 92 gods and 88 demons fighting for the elixir of immortality and a snake caught up in the middle. The gods hold the tail, the demons hold the head, while the snake coils itself around Mt. Mandala. Each time the gods and demons pull from their sides, the mountain turns and the ocean churns.

There is a lot of static (translational) symmetry in the repeating figures. The symmetry is also broken (the head and tail of the world-snake, gods and demons, 92 vs. 88). But ultimately there is a dynamical balance, at least for the moment.

And to close the circle (another symmetry), my new “home” town is famous for its own version of a building in a dynamic equilibrium, that is inspired by dancers: The Dancing House.




von Neumann’s Nightmare

Via the always excellent Stephen Hsu – apparently the concept of Technological Singularity can be traced back to von Neumann’s nightmare:
One night in early 1945, just back from Los Alamos, von Neumann woke in a state of alarm in the middle of the night and told his wife Klari:

“… we are creating … a monster whose influence is going to change history … this is only the beginning! The energy source which is now being made available will make scientists the most hated and most wanted citizens in any country.

The world could be conquered, but this nation of puritans will not grab its chance; we will be able to go into space way beyond the moon if only people could keep pace with what they create …”

He then predicted the future indispensable role of automation, becoming so agitated that he had to be put to sleep by a strong drink and sleeping pills. Source: Von Neumann, Morgenstern, and the Creation of Game Theory: From Chess to Social Science, 1900-1960.

In his obituary for John von Neumann, Ulam recalled a conversation with von Neumann:

[about the] “ever accelerating progress of technology and changes in the mode of human life, which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue.”

It is only fitting that the (co-)father of both game theory and computation also “discovered” their common endpoint.

As you recall the concept of intelligence explosion is attributed to I. J. Good and the term technological singularity to Vernor Vinge.

Authentic and Inauthentic Loyalty

Loyalty is a desirable core value, but I think, that sometimes it is misunderstood. I’d wish to distinguish between two types of loyalty – authentic and inauthentic.

Authentic loyalty is the stance one thinks of when we talk about loyalty – the steadfast quality of sticking with a person, group or organisation, that comes from a deep feeling of caring and belonging.

Inauthentic loyalty looks on the surface the same – grin and bear, stick through good and bad times, . Under the surface, it can however harbor resentment, contempt and even anger.

If you consider loyalty as your core value, it is worth investigating what is its ultimate source.

 The near enemy of loyalty is complacency

Buddhism has an interesting concepts of “near enemies”. These are undesirable emotions that can be confused with noble ones. For example, compassion is a desirable emotion and its near enemy is pity (because pity is is divise at its core).

I’d suggest that near enemy is of loyalty is compacency.

What on looks like steadfast loyalty, can under the surface be complacency based on fear, lazyness or other source of resistance to change. This “poisons” the relation and causes slow, silent damage to both sides (individuals, groups, organisations).

Can inauthentic loyalty turn into authentic?

Sometimes “fake-it till you make-it” does indeed work. My suspicion is that inauthentic loyalty can indeed sometime turn into the deep, authentic stance. However, I expect this is rare in personal relationships and pretty much non-existent towards organisations and business.

Resentment and contempt are particularly hard emotions to deal with and time tends to deepen them rather than heal.

So if the way “through” is unlikely, what other possibility can we have?

Escaping the trap of inauthentic loyalty

As with any “near enemy”, inauthentic loyalty can only be recognized with patient, self-compassionate, introspective mindfulness. The recognition is often not a singular moment a breakthrough, but a slow dawning, possibly painful and with backtracks.

Ultimately moving beyond inauthentic loyalty brings liberation and eventually growth, but it is not an easy process.


Few notes on crony beliefs

Kevin Simler at Melting Asphalt recently published a very nice summary post – Crony Beliefs. Building on the evo-psych work of Trivers, Haidt and Kurzban he distinguishes between two types of beliefs (“employees”):

  • Meritocratic Beliefs
    • beliefs that are entertained because of their epistemic value, i.e. beliefs that pay their rent in accurate predictions about the world
  • Crony Beliefs
    • beliefs that are not accurate representations of reality, but are kept around because of social value and signaling purposes

The important thing to understand is that Crony Beliefs pay rent too – but not in accurate predictions, but rather as entry and maintenance costs of social bonds.

This observation fits very well with the “bleeding-heart” approach to cognitive biases: while our reasoning is deeply flawed, most heuristics are understandable from the evolutionary point of view, actually work quite often (especially in the simpler ancestral environments) and, as Kevin adds, can even provide a lot of utility, albeit non-epistemic.

There are only few (minor) things I’d like to add to this picture.

Break-downs of meritocracy still exists

From Kevin:

I, for one, typically explain my own misbeliefs (as well as those I see in others) as rationality errors, breakdowns of the meritocracy. But what I’m arguing here is that most of these misbeliefs are features, not bugs. What looks like a market failure is actually crony capitalism. What looks like irrationality is actually streamlined epistemic corruption.

My feeling is that despite this, true meritocracy breakdowns are still more common than true crony beliefs. While there are a LOT of social biases (authority bias, bias from social desirability, conformity bias, groupthink etc.), I think still most of the biases are simply heuristics extrapolating too far & breaking down, i.e. providing neither epistemic nor social value.

Bounded-rationality, compartmenalization and reflective equilibrium

AND ON TOP of that we have a bunch of beliefs that don’t have any causal links leading outside of our skulls, and beliefs that we have, but kind-of-sort-of don’t really know if they are useful in any way, because the slow combine-harvester of System 2 (or some other possibly non-conscious process) didn’t yet mowed over the belief and “decided” to integrate it, throw it away, or leave around unattached.


I’d really like to learn more about how this bounded belief examination and integration works.

Mindsets vs. beliefs

Finally a point that Julia Galef raised: beliefs are possibly not quite stable and the same metaphor could be applied to mindsets instead.

Instead of crony/meritocratic employees (beliefs), you might have a crony/meritocratic HR & hiring process.

Julia’s metaphor of scout vs. soldier mindset roughly maps to meritocratic vs. crony beliefs too.

Kevin’s article is definitely worth reading in its entirety, because it gives a very vivid metaphor to understand the interplay of the two types of beliefs.