I am a generalist quantitative researcher. I am open to volunteering and paid work. I welcome suggestions for posts. You can give me feedback here (anonymously or not).
Vasco Grilošø
Karma: 11,784
Hi Laura. That finding also jumped out to me. I suspect orgs are less willing to improve MEL than suggested by that quote. ā77% [of orgs] say funders rarely or never fund MEL (excl. n/āa results)ā, but I assume this refer to funding restricted to MEL, whereas what I think matters to assess orgsā preferences is whether there are funds which can be used to improve MEL. I would be surprised if funders restricted funding to non-MEL activities. So I believe orgs could spend more on MEL if they wanted to.
ā75% of organizations say improving MEL is important or urgentā, but I find it difficult to interpret this without knowing what orgs would say about other activities. Maybe the vast majority of orgs considers more important or urgent improving other activities, which could explain why āonly 18% would definitely pay for itā.
Also, in regards to Bobās book, it seems like it would make sense, rather than adding the modelās estimates together to instead act on each model separately, i.e. assigning 5% of resources to help animals according to the equality model, 25% according to the neurophysiological model, and 70% according to the proxies model. Then, within the 70% for the proxies model, we could assign 33% of resources to applying a high exponent to the welfare ranges, 33% to applying no exponent to the welfare ranges, and 33% to applying a low exponent to the welfare ranges.
Something like this is proposed in the article āMoral Uncertainty, Proportionality and Bargainingā.
Only physical consciousness is falsifiable? I think so. So I believe it is better to focus on physicalist theories of consciousness. Relatedly, you may be interested in this discussion between me and Wladimir Alonso from the Welfare Footprint Institute (WFI) about whether sentience is binary.
Hi Gregory. Great post.
I mentioned before that IRT is basically an Elo for the game of responding correctly to benchmark questions. For Elo, hereās the win probability of player A vs. player B, given respective strengths Sa and Sb:
P(A wins) = 1 /ā [1 + 10^((SbāSa)/ā400)]
Nitpick. I think the above is the probability of winning plus half of the probability of drawing.
I broadly agree with the points you made in that comment.
You may be interested in my estimates of the total welfare of various organism populations assuming welfare per fully-happy-animal-year is a power law of the individual number of neurons, or of the basal metabolic rate (BMR) at 25 ĀŗC. Here is some context about why a power lay may be appropriate.
Hi Aidan and Aaron. Great news. The program sounds very promising.
Meanwhile, the animal movement needs more new interventions and organizations in the pipeline so that efforts to fix factory farming can scale in the coming years.
I like that you say āfixā instead of āendā factory farming. Fixing it is more aligned with increasing animal welfare, as there is large uncertainty about which animals have positive or negative lives (worth living or not from their own perspective).
Thanks, Anthony.
I think it would help a lot if you spelled out the premises more, because theyāre quite opaque to me as written. E.g. I donāt know what āthe norm reveals a choice functionā means. (I think this kind of use of jargon without giving context is a common failure mode of current LLM summaries.)
I asked Claude to update the post to address your comment. There is now a section with the setup of the theorems, and clearer premises. Are they sufficiently understandable now?
Also, if I understand correctly, ābehaviour maximizes a family of preference orderings...ā means that the result only shows that we can represent an agentās behavior as satisfying completeness.
I only briefly skimmed the article, but I agree with your interpretation. Claude agrees too.
The completeness here is a property of a ranking reconstructed from choices, and it comes almost entirely from premise 0: because the rule always names something acceptable, every pair gets settled. [This is now clarified in the section of the linkpost with the setup.] But the rule may name both options as acceptable, which the theorem records as the two being equally good ā and that is also exactly how an agent who found them incomparable, and picked arbitrarily, would behave. The result therefore cannot distinguish āequally goodā from ānot comparableā, and so does not show that an agent deliberating about what to do must arrive at a complete ranking. It shows their behaviour is representable as if they had one.
Hi James.
This question gets to me a lot. It seems totally unclear how much the valence of an experience correlates to behavior. Like, it seems plausible that extraordinarily little pain could motivate behavior in simple organisms, but itās also hard to imagine, since, as a complex organism, I need pretty intense pain to be motivated to significantly change my behavior. And, it doesnāt seem like thereās any way we could reduce uncertainty about this.
Makes sense. Relatedly, you may be interested in the post āDo primitive sentient organisms feel extreme pain? disentangling intensity range and resolutionā by Wladimir Alonso and Cynthia Schuck.
Also, in regards to Bobās book, it seems like it would make sense, rather than adding the modelās estimates together to instead act on each model separately, i.e. assigning 5% of resources to help animals according to the equality model, 25% according to the neurophysiological model, and 70% according to the proxies model. Then, within the 70% for the proxies model, we could assign 33% of resources to applying a high exponent to the welfare ranges, 33% to applying no exponent to the welfare ranges, and 33% to applying a low exponent to the welfare ranges.
I think the above goes against maximising expected value in principle, even if it may be a good approach in practive, and I believe maximising expected value follows from very intuitive premises.
The problem of what percentages you should use and what models to use would, of course, still remain though.
Right.
I think consistently applying your arguments suggest you should assign 0 probability in these cases, and so your models blow up in complexity and you become too credulous, contrary to Occamās razor.
There is also Hitchensās razor.
What can be asserted without evidence can also be dismissed without evidence.
Readers of this post may be interested in this summary of the setup, premises, and conclusions of the theorems presented in Consequentialist Foundations for Expected Utility by Peter J. Hammond.
I just linkposted a summary of the setup, premises, and conclusions of the theorems presented in Consequentialist Foundations for Expected Utility by Peter J. Hammond.
ConĀseĀquenĀtialĀist FounĀdaĀtions for ExĀpected Utility
David Spiegelhalter and Mike Pearson calculated driving a car in the UK for 250 mi increases oneās risk of death by 1 micromort, which is 2.48*10^-7 per 100 km.
Motor vehicle road injuries (excluding motorcyclistsā) caused 22.1 M years of life lost, and 436 k deaths in 2019, which imply 50.7 years of life lost per death from motor vehicle road injuries in 2019. This times the increase in the risk of death due to driving a car in the UK equals 6.62 person-min less of life expectancy per 100 km.
There were 2.7 road fatalities per 1 billion miles driven by cars in Great Britain in 2024 (see Table 4), 1.68*10^-7 road fatalities per 100 km (= 2.7/ā(1*10^9*1.61)*100), 67.7 % (= 1.68*10^-7/ā(2.48*10^-7)) as much as I estimated above. As a result, for the above 50.7 years of life lost per death, there is a reduction in life expectancy per 10 km of driving a car of 4.48 min (= 0.677*6.62).
Thanks. Relatedly, you may be interested in the comments from me and Wladimir on this post from Benthamās Bulldog arguing for the possibility of intense agony in many species.
Hi Mo.
I do think many of your points are correct. The strongest argument to your post title I can think of is sparse evidence + multiple models and youāre uncertain between them ā precise priors are unwarranted ā use an interval instead, possibly quite wide like [10ā»ā¶, 0.5] (just to make something up) ā updating may not collapse this wide interval ā so EV-maxxing becomes undefined ā so switch to other decision criteria, e.g. maybe robustness to harms, which is less Bayesian as per your post title ā choose robustly good actions, maintain option value, build capacity etc. Which is basically what most meta interventions are about, no?
This argument works with sharp probabilities too? If the distributions for the cost-effectiveness are very wide, the expected cost-effectiveness of decreasing uncertainty or building capacity would tend to be higher than the highest expected cost-effectiveness of the interventions under evaluation, even if these are all sharp values?
I do not want to give up completeness because it follows from 3 super intuitive premises.
Behaviour norms are considered for decision trees which allow both objective probabilities and uncertain states of the world with unknown probabilities. Terminal nodes have consequences in a given domain [premise 1; unrestricted domain]. Behaviour is required to be consistent in subtrees [premise 2; dynamic consistency]. Consequentialist behaviour, by definition, reveals a consequence choice function independent of the structure of the decision tree [premise 3; consequentialism]. It implies that behaviour reveals a revealed preference ordering [āa complete, transitive, binary relationā] satisfying both the independence axiom and a novel form of surething principle.
Hi James. Nice points.
Reason #6: People rarely offer extraordinarily low probabilities.
Whenever I ask someone how likely they think something is, they pretty much never give a probability less than .1% unless that something is religious in nature. Given this, it seems like people systematically over estimate low probabilities because they fail to consider probabilities such as 10^-7 or 10^-53.
Some people would argue models predicting a probability of at least 0.1 % should have significant weight, for example, at least 10 %, which would imply an expected probability of at least 0.01 % (= 1*10^-3*0.1). However, I have significant concerns about this kind of reasoning. I worry the weights of the models are close to arbitrary. For instance, in Bob Fischerās book about comparing welfare across species, there seems to be only 1 line about the weights. āWe assigned 30 percent credence to the neurophysiological model, 10 percent to the equality model, and 60 percent to the simple additive modelā. People usually give weights that are at least 0.1/āānumber of modelsā, which is at least 3.33 % (= 0.1/ā3) for 3 models, when it is quite hard to estimate the weights. However, giving weights which are not much smaller than the uniform weight of 1/āānumber of modelsā could easily lead to huge mistakes. As a silly example, if I asked random people with age 7 about whether the gravitational force between 2 objects is proportional to ādistanceā^-2 (correct answer), ādistanceā^-20, or ādistanceā^-200, I imagine I would get a significant fraction picking the exponents of ā20 and ā200. Assuming 60 % picked ā2, 20 % picked ā20, and 20 % picked ā200, one may naively conclude the mean exponent of ā45.2 (= 0.6*(-2) + 0.2*(-20) + 0.2*(-200)) is reasonable. Yet, there is lots of empirical evidence against this which the respondants are not aware of. The right conclusion would be that the respondants have practically no idea about the right exponent because they would not be able to adequately justify their picks.
Could ConĀsciousĀness Be an IlluĀsion?
Hi Cat. Thanks for tagging me.
Hi Necta. You may be interested in joining EA Lisbonās WA group. There are some people in the group like me who are interested in increasing animal welfare. We have monthly meetups. You may also be interested in the meetups from the groups Lisbon Veggie Friends and Lisbon Sustainability Community.
Hello Hazem. You can try to turn it into motivation to publish a post (including a linkpost/ācrosspost) for each untitled draft.
Hi Kieran. Welcome to the EA Forum. That matches my read.