Meet on the ground floor of QNC, on or near the gray slab benches at 7:20ish, and a regular will let everyone into a room in IQC for 7:30 pm. You can see the precise location on plus.codes. The benches are near the Ring Road entrance, and across the hall from an orange-tinted lab space.
As always, vegetarian snacks will be provided.
Discussion
Hello fellow utility maximizers.
Would you take a 50/50 bet to double or halve your money? Would you take a 50/50 bet where you win $5 at the risk of losing $4? What if you'd win $10,000 or lose $10?
In this week's meetup we will refresh our understanding of marginal utility and its implications for risk management.
If you missed Utility Maximizing 101, here are the course notes.
1. Would you take a 50/50 bet to double or halve your money?
2. Does the fact that utility maximizers are approximately risk-neutral match with your intuitions about rational decision making?
3. Given the results in the readings, do you think you are rationally risk averse?
4. Bernoulli said it is "Nature’s admonition to avoid the dice altogether"; gambling is generally taboo across a wide spectrum of human culture. Is there a reason for this taboo that isn't captured in the analysis of the readings?
5. Do you find the independence axiom unintuitive?
6. Some of the comments on Yudkowsky's post point out that the paradox can be resolved by including the utility of the feeling of certainty in choice 1A). Do you think that this is a satisfactory resolution of the paradox?
7. What evolutionary advantage (if any) do you think human bias towards risk avoidance served? Is that advantage still applicable in modern society?
Meet on the ground floor of QNC, on or near the gray slab benches at 7:20ish, and a regular will let everyone into a room in IQC for 7:30 pm. You can see the precise location on plus.codes. The benches are near the Ring Road entrance, and across the hall from an orange-tinted lab space.
As always, vegetarian snacks will be provided.
Discussion
Hello fellow utility maximizers.
Would you take a 50/50 bet to double or halve your money? Would you take a 50/50 bet where you win $5 at the risk of losing $4? What if you'd win $10,000 or lose $10?
In this week's meetup we will refresh our understanding of marginal utility and its implications for risk management.
If you missed Utility Maximizing 101, here are the course notes.
Primary Readings
1. The Allais Paradox, Yudkowsky
https://www.lesswrong.com/posts/zJZvoiwydJ5zvzTHK/the-allais-paradox
2. Anomalies: Risk Aversion, Rabin & Thaler
https://www.aeaweb.org/articles?id=10.1257/jep.15.1.219
3. Daniel Bernoulli's Utility, Ben Lynn
https://crypto.stanford.edu/~blynn/pr/utility.html
Auxiliary Readings:
a. How Economists Came to Accept Expected Utility Theory: The Case of Samuelson and Savage, Moscati
https://www.aeaweb.org/articles?id=10.1257/jep.30.2.219
b. The Kelly Criterion, Zvi
https://www.lesswrong.com/posts/BZ6XaCwN4QGgH9CxF/the-kelly-criterion
c. The Allais "Paradox" and Scam Vulnerability, Karl Hammer
https://thefutureprimaeval.net/the-allais-paradox-and-scam-vulnerability/
Discussion Questions
1. Would you take a 50/50 bet to double or halve your money?
2. Does the fact that utility maximizers are approximately risk-neutral match with your intuitions about rational decision making?
3. Given the results in the readings, do you think you are rationally risk averse?
4. Bernoulli said it is "Nature’s admonition to avoid the dice altogether"; gambling is generally taboo across a wide spectrum of human culture. Is there a reason for this taboo that isn't captured in the analysis of the readings?
5. Do you find the independence axiom unintuitive?
6. Some of the comments on Yudkowsky's post point out that the paradox can be resolved by including the utility of the feeling of certainty in choice 1A). Do you think that this is a satisfactory resolution of the paradox?
7. What evolutionary advantage (if any) do you think human bias towards risk avoidance served? Is that advantage still applicable in modern society?
Posted on: