DominikPeters

Karma: 162

DominikPeters 17 Apr 2024 21:51 UTC
13 points
0 ∶ 0
on: Future of Humanity Institute 2005-2024: Final Report
From Bostrom’s website, an updated “My Work” section reads:
… That’s why I founded the Future of Humanity Institute at Oxford University in 2005. FHI brought together an interdisciplinary bunch of brilliant (and eccentric!) minds, and sought to shield them as much as possible from the pressures of regular career academia; and thus were laid the foundations for exciting new fields of study.

Those were heady years. FHI was a unique place—extremely intellectually alive and creative—and remarkable progress was made. FHI was also quite fertile, spawning a number of academic offshoots, nonprofits, and foundations. It helped incubate the AI safety research field, the existential risk and rationalist communities, and the effective altruism movement. Ideas and concepts born within this small research center have since spread far and wide, and many of its alumni have gone on to important positions in other institutions.
Today, there is a much broader base of support for the kind of work that FHI was set up to enable, and it has basically served out its purpose. (The local faculty administrative bureaucracy has also become increasingly stifling.) I think those who were there during its heyday will remember it fondly. I feel privileged to have been a part of it and to have worked with the many remarkable individuals who flocked around it. [Update: FHI officially closed on 16 April 2024]

DominikPeters 10 Nov 2023 14:37 UTC
17 points
2 ∶ 0
in reply to: Lizka’s comment on: Lizka’s Shortform
I’m a researcher on voting theory, with a focus on voting over how to divide a budget between uses. Sorry I found this post late, so probably things are already decided but I thought I’d add my thoughts. I’m going to assume approval voting as input format.
There is an important high-level decision to make first regarding the objective: do we want to pick charities with the highest support (majoritarian) or do we want to give everyone equal influence on the outcome if possible (proportionality)?
If the answer is “majoritarian”, then the simplest method makes the most sense: give all the money to the charity with the highest approval score. (This maximizes the sum of voter utilities, if you define voter utility to be the amount of money that goes to the charities a voter approves.)
If the answer is “proportionality”, my top recommendation would be to drop the idea of having only 3 winners and not impose a limit, and instead use the Nash Product rule to decide how the money is split [paper, wikipedia]. This rule has a nice interpretation where let’s say there are 100 voters, then every voter is assigned 1/100th of the budget and gets a guarantee that this part is only spent on charities that the voter has approved. The exact proportions of how the voter share is used is decided based on the overall popularity of the charities. This rule has various nice properties, including Pareto efficiency and strong proportionality properties (guaranteeing things like “if 30% of voters vote for animal charities, then 30% of the budget will be spent on animal charities”).
If you want to stick with the 3 winner constraint, there is no academic research about this exact type of voting situation. But if proportionality is desired, I would select the 3 winners as not the 3 charities with the highest vote score, but instead use Proportional Approval Voting [wikipedia] to make the selection. This would avoid the issue that @Tetraspace identified in another comment, where there is a risk that all 3 top charities are similar and belong to the largest subgroup of voters. Once the selection of 3 charities is done, I would not split the money in proportion to approval scores but either (a) split it equally, or (b) normalize the scores so that a voter who approved 2 of the 3 winners contributes 0.5 points to each of them, instead of 1 point to each. Otherwise those who approved 2 out of 3 get higher voting weight.
I’m happy to discuss further.

DominikPeters 27 Jul 2023 7:01 UTC
2 points
2 ∶ 0
in reply to: Arturo Macias’s comment on: Underwater Torture Chambers: The Horror Of Fish Farming
I don’t think that’s right. Half the fishes sold to humans for consumption are farmed, but if I understand correctly, a lot of caught fishes are used as feed in farms. Those fishes used for feed will also be relatively smaller and therefore more numerous. So I expect it to indeed be an order of magnitude error.

DominikPeters 26 Jul 2023 21:50 UTC
2 points
0 ∶ 0
in reply to: Arturo Macias’s comment on: Underwater Torture Chambers: The Horror Of Fish Farming
As far as I can tell, “3-8 billion killed fish” figure indeed refers to caught fishes, not to farmed fishes (following the links it comes about from the classic 3 trillion / year estimate; http://fishcount.org.uk/published/std/fishcountstudy.pdf). I think that this part of the article is misleading; the relevant number would be the total count of fishes living in farms at any given time, as well as the number of farm slaughters.

DominikPeters 27 Feb 2023 22:09 UTC
16 points
6 ∶ 0
in reply to: Bella’s comment on: 80,000 Hours has been putting much more resources into growing our audience
Interesting that the 80k book is so popular. Is this the book that came out in late 2016, or has the content been updated? If not, it may be worth doing a second edition, since I assume that 80k’s thinking has evolved significantly since publication.

DominikPeters 17 Nov 2022 7:44 UTC
54 points
29 ∶ 10
in reply to: Jeff Kaufman’s comment on: Kelsey Piper’s recent interview of SBF
Kelsey’s messages are written in a style of informality that strongly suggests a casual conversation with a friend, and not a formal interview with a journalist. The emoji reactions have a similar effect, and there isn’t an introductory message along the lines of “would you be happy to talk to vox”. This overall seemed somewhat manipulative to me.

DominikPeters 13 Nov 2022 19:04 UTC
19 points
2 ∶ 0
in reply to: Ozzie Gooen’s comment on: A personal statement on FTX
I believe this is SBF’s blog: http://measuringshadowsblog.blogspot.com/

DominikPeters 1 Jul 2022 9:09 UTC
19 points
0 ∶ 0
on: Future Fund June 2022 Update
“Our sense is we’re able to generate >2x more value per time with our other activities [than with open calls]”: does this number include an estimate of the time spent by regrantors? (Even if it doesn’t, the 2x figure is still interesting.)

DominikPeters 5 Feb 2022 17:02 UTC
22 points
0 ∶ 0
on: New EA Cause Area: Run Blackwell’s Bookstore
The publishing arm was sold to Wiley for £572 million, so it probably will be difficult to buy it back. I would guess that the deal would have included something like a transfer of a trademark, which would make it difficult to restart publishing afresh under the name Blackwell’s.

DominikPeters 2 Jan 2021 21:22 UTC
2 points
0 ∶ 0
in reply to: EdoArad’s comment on: 10 Habits I recommend (2020)
For zoom, when scheduling a meeting and marking it as “recurring”, the link should stay valid indefinitely.

DominikPeters 23 Mar 2018 13:47 UTC
0 points
0 ∶ 0
in reply to: Dunja’s comment on: How effective and efficient is the funding policy of Open Philanthropy concerning projects on AI risks?
Here’s a link: http://eptcs.web.cse.unsw.edu.au/paper.cgi?TARK2017.16

DominikPeters 28 Feb 2018 10:26 UTC
4 points
0 ∶ 0
in reply to: Gregory Lewis’s comment on: How effective and efficient is the funding policy of Open Philanthropy concerning projects on AI risks?
Also, parts of their logical induction paper were published/presented at TARK-2017, which is a reasonable fit for the paper, and a respectable though not a top conference.

DominikPeters 27 Oct 2017 7:50 UTC
1 point
0 ∶ 0
in reply to: Milan_Griffes’s comment on: Rob Wiblin’s top EconTalk episode recommendations
Not sure about iTunes/iOS; probably I’d need to submit the podcast to Apple for approval which I don’t have enough permission to do :) Maybe there are non-Apple-protected apps? Or switch to Android.

DominikPeters 26 Oct 2017 21:54 UTC
1 point
0 ∶ 0
in reply to: Milan_Griffes’s comment on: Rob Wiblin’s top EconTalk episode recommendations
I’ve made a feed with Wiblin’s top 10 episodes for easy importing into podcast apps.

http://bit.ly/econtalk-wiblin

expands to https://dl.getdropbox.com/s/hjdlhtv6xtklhxv/econtalk-wiblin.xml

DominikPeters 3 Aug 2017 20:11 UTC
3 points
0 ∶ 0
on: How much donation splitting is there, and should you split?
Under certain circumstances, having moral uncertainty over theories that are purely ordinal may lead to the recommendation to split. Example: Suppose there are three charities A,B,C, and four options: donating 100% to one of A, B, C, or splitting the money equally between them (which we will call S). Let’s ignore other ways of splitting. Suppose you have equal credence of 33% in three different theories:

1: A > S > B > C

2: B > S > C > A

3: C > S > A > B

Given the ranking over charities, it is rational in something like a von Neumann-Morgenstern sense to rank S second. But with these theories and these credences, one can see that S is the Condorcet winner and it is also the unique Borda winner, so that S would be uniquely recommended by essentially all voting rules, including Borda, the system favoured by Will MacAskill. In this example, contrary to the example in the OP, option S is not Pareto-dominated by another option, so that the unanimity principle does not bite.

This example crucially depends on only having ordinal information available, since with cardinal information (and expected value maximisation) we would never uniquely recommend splitting, as Tom notes, and so I don’t think the argument in favour of splitting from moral uncertainty is particularly strong or robust.