Overcoming Bias is a group blog on the systemic mistakes humans make, and how we can possibly correct them. The primary contributors are Robin Hanson of George Mason University and Eliezer Yudkowsky of the Machine Intelligence Research Institute. Common topics include "cognitive psychology, evolutionary psychology, microeconomics, applied statistics, social psychology, probability and decision theory, even a bit of Artificial Intelligence now and then."
The complete list of Yudkowsky's posts on Overcoming Bias was compiled by Andrew Hay here.
A categorized list of Yudkowsky's posts moved from Overcoming Bias to Less Wrong is available at Less Wrong/All Articles.
Title | Author | Date | Summary |
|---|---|---|---|
2006-11-20 | Description of OB and how to contribute. | ||
2006-11-20 | Should we teach children about self-interest explanations and sociobiology earlier? | ||
2006-11-21 | Discussion of Philip Tetlock's Fox/Hedgehog classification guide in Expert Political Judgement. | ||
2006-11-21 | Is there a bias towards working hard and against spending enough time with family? | ||
2006-11-21 | Biases may exist on an individual level, even if they cancel out on a group level, so even apparently contradictory bromides might highlight important types of failure. | ||
2006-11-22 | Some opinions are highly heritable, so put extra scrutiny on those beliefs. | ||
2006-11-22 | Rationality as martial art. Individuals should be able to train their mind like they train muscles. | ||
2006-11-23 | First known example of a market designed primarily to gain information from was created by Xanadu, Inc. in 1990. | ||
2006-11-24 | Ratio of Democrats to Republicans in academia is 5:1 compared to roughly 1:1 in general populace. Is this due to intelligence and information, or social reasons? | ||
2006-11-24 | Students admit to cheating, lying, and theft, but 75% think they are more ethical than their peers. | ||
2006-11-25 | "A bias is a non-rational factor that systematically pushes one's beliefs in some domain in one direction." | ||
2006-11-26 | According to Philip Tetlock, foxes (a flexible, tentative cognitive style) are more successful than at forecasting. Hedgehogs do worse than a random guess. | ||
2006-11-26 | Even though bias might have a broader technical meaning, it is better to think of it as "cheaply avoidable error". | ||
2006-11-26 | Paternalism to correct common biases and public choice considerations. | ||
2006-11-26 | We seek the truth for intellectual curiosity, pragmatic reasons, and for its own sake, although there is danger in thinking a moral duty to be rational exists. | ||
2006-11-26 | A bias is an obstacle to us knowing the truth. Biases are best defined by observed patterns of errors, not by an actual definition, because there are so many ways to be wrong. | ||
2006-11-27 | Publicizers are often better known for a discovery than the actual innovator. | ||
2006-11-27 | Areas with lower population density tend to be friendlier, contrary to popular belief. | ||
2006-11-28 | Bryan Caplan's work on voter irrationality identifies correlated errors, not biases. Public more skeptical of economics than physics experts. | ||
2006-11-28 | Attempts to reduce bias feasible, even though our cognition is limited. The desire to deduce bias is not itself a bias. | ||
2006-11-28 | Random error helps "evolutionary" development of truth. Bias needed to fight bias. Bias might generate beneficial self-fulfilling prophecies. Errors needed to exercise reason and debating skill. | ||
2006-11-28 | Disagreement in the philosophy of disagreement. | ||
2006-11-30 | What do complaints by women signal? | ||
2006-11-30 | New security might function as a pure prediction market alternative. | ||
2006-12-01 | Everyone knows local sport reporting is biased, but no one cares. What about other biases if they become widely known? | ||
2006-12-01 | Use humility to justify further action, not as an excuse for laziness, irrationality, or ignorance. Humility can be too easy to admit to. | ||
2006-12-02 | Conflicts between good story telling and good forecasting. | ||
2006-12-02 | Self-deceptive illusions appear to make us happier. | ||
2006-12-03 | Traders who make the most money in a prediction market might not be the best forecaster. | ||
2006-12-03 | Time-inconsistent preferences and akrasia | ||
2006-12-04 | Seen biases might cancel out unseen ones, so partial debiasing might make us less accurate, although avoid using this as an excuse. | ||
2006-12-05 | Bosses appear to interpret accurate estimates as a signal of incompetence. | ||
2006-12-05 | What does it mean if a study finds the effect of a small increase in the minimum wage to be near zero? | ||
2006-12-06 | Some disagreements might be reasonable because of incommunicable evidence and insights. | ||
2006-12-06 | |||
2006-12-07 | |||
2006-12-08 | |||
2006-12-09 | |||
2006-12-10 | |||
2006-12-10 | |||
2006-12-11 | |||
2006-12-11 | |||
2006-12-11 | |||
2006-12-12 | |||
2006-12-12 | |||
2006-12-13 | |||
2006-12-14 | |||
2006-12-14 | |||