Recent searches

No recent searches

Search options

Only available when logged in.
sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!

Administered by:

Server stats:

587
active users

sigmoid.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.4.3

#gametheory

0 posts0 participants0 posts today
James

Goodhart's Law is one of those ideas, once you learn it, you find it everywhere. I rewrote this post in order to read it for the podcast (so it flowed more) and it turned out so much better that I've updated the post. #MentalModels #politics #philosophy #GameTheory #GreatIdeas #Goodhart #Economics

GOODHART'S LAW

nonzerosum.gamesGOODHART'S LAWGoodhart's Law illustrates how measures can become unreliable targets, leading to unintended consequences, much like moths drawn to harmful lights.
Vince Knight (He/Him)

I have been playing around with auction theory, I put sold together which lets you simulate auctions:

github.com/drvinceknight/sold

Here is some code to simulate a second pay auction where 1 of the bidders shades their bid (optimal behaviour in a first pay auction) and the other bids their true value.

The true value bid is known to be weakly dominant in a second pay auction) which is kind of what is confirmed by the plot.

#gametheory#auction
Vince Knight (He/Him)

A few of us have put together a #gametheory discord server.

We're not yet entirely sure what it's going to be like so please do join and help shape it if you're interested (even slightly) in Game Theory.

> We aim to bring together hobbyists, educators, students, researchers, and developers to learn, share, and collaborate across all areas of game theory.

Here is the invite link: discord.gg/NfTAkhAeyc

(Boosts of this to relevant people would be appreciated :))

DiscordJoin the Equilibrium Explorers Discord Server!Game Theory server — whether you're just starting out, teaching it, researching it, or applying it in code or policy. | 5 members
*
ma𝕏pool

Polymarket question: "will Jesus Christ return in 2025?" has a 3% probability.

Polymarket wants to provide a public good: correct probability estimates. It fails because how the market is structured and limited.

spoiler: "The Yes people are betting that, later this year, their counterparties (the No betters) will want cash (to bet on other markets), and so will sell out of their No positions at a higher price."

ericneyman.wordpress.com/2025/

Unexpected Values · Will Jesus Christ return in an election year?Thanks to Jesse Richardson for discussion. Polymarket asks: will Jesus Christ return in 2025? In the three days since the market opened, traders have wagered over $100,000 on this question. The mar…
#economics#predictionMarkets#mechanismDesign
@gurupanguji

🔗 Most tech companies and platforms really don’t know what people like or why…

From the head of design at Substack,

I am very confident this is true of TikTok today. TikTok is an amazing product because of its format and its ranking (and how they relate). There are possibly “experts on psychology” in their buildings, writing long emails no one reads, but TikTok doesn’t know the first thing about “how to control people” or “who you are at a deep level” or “the ways to exploit the limbic system” or anything else along those lines; what they have is: vast and fresh inventory and incredible ranking, especially their explore / exploit balance. TikTok executives cannot predict what their children or spouses will do, let alone what you or the rest of the American people will do. They control and understand far less than people suspect.

The irrepressible monkey in the machine – by Mills Baker

While the specific post itself is a reaction to another incredible post by Frank Lantz, this part rang true to me. The other bit that resonated was this:

Inside companies I’ve worked at, the only way to describe the relationship to users and customers is to say it was fearful. Companies are desperate to appeal to the market; companies are terrified of their users, whom they do not really understand.

While the state was not always fear / terror, there’s a kernel of truth to this observation.

It’s also hilarious to imagine that the tech platforms really have the capability to influence the people to consider their brand or a competitor’s brand in any way. The core of the reason for this is inherently an iterative game.

This competitive dynamic is omnipresent. Even if some Ph.D figured out “how humans work” in some sense that enabled a company to exploit it, humans would very quickly incorporate this implicit or explicit knowledge into their thinking and the hack would stop working.1

The point is merely that there is no permanent hack to be found in any of these fields, or in the academic work they ostensibly make use of. Humans adapt, change, shift, and remain just outrageously hard to influence, let alone “control,” and this cannot change, no matter how expert the efforts at manipulation are.

substack.comMills Baker | SubstackHead of Design at Substack. "It is better to say 'I am suffering' than to say 'This landscape is ugly.'" - Simone Weil. From, in, and of New Orleans.
#design#gameTheory#philosophy
TrendingLive feeds
About