Signal and Noise: Why So Many Predictions Fail – but Some Don't

By D. Subedi. 2013

469 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience

View Paper: https://www.semanticscholar.org/paper/f5dc7d82b9e51ae8c428b2de1965a0ab5b3354c5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cognitive Neuroscience

Citations in the corpus

Testing the ability of the surprisingly popular method to predict NFL games (2018, background): “Our results suggest that it may be useful in forecasting applications involving genuine predictions, as an alternative or complementary method to more elaborate approaches like prediction markets (Silver, 2012).”
Accounting for Epistemic Uncertainty in PSHA: Logic Tree and Ensemble Modeling (2015, background): “Nate Silver’s recent book (Silver, 2012) gives a wide range of successful applications and uses of ensemble modeling.”
Machine learning bandgaps of double perovskites (2016, background): “…phenomenal success in the fields of cheminformatics 10 , game theory 11 , pattern recognition 12 , artificial intelligence 13 , event forecasting 14 etc. and are now being customized for materials informatics to help identify next generation materials break-throughs and process optimizations…”
Personality Strength and Situational Influences on Behavior (2015, background): “For example, because they tend to learn from feedback and understand the complexities engendered by situational variability, weak personalities (“foxes”) are likely to be more accurate forecasters than strong personalities (“hedgehogs”; Silver, 2012).”
Asiaphoria Meets Regression to the Mean (2014, background): “Silver (2012) argues the models being used by the risk-rating agencies underestimated the default risk of some bonds not by percentage points or even a single order of magnitude, but by a factor of 200.! !”
The Fallacy of AI Functionality (2022, background): “The challenge of dealing with an influx of fraudulent or dysfunctional products is one that has plaguedmany industries, including food safety [24], medicine [12, 17], financial modeling [170], civil aviation [87] and the automobile indus-”
When do people rely on algorithms (2016, background): “, 2005) and talent scouts (Silver, 2012) previously discussed, reject algorithms that try to do their jobs.”
Algorithm aversion: people erroneously avoid algorithms after seeing them err. (2015, background): “Thus, across the vast majority of forecasting tasks, algorithmic forecasts are more accurate than human forecasts (see also Silver, 2012).”
Psychotherapy in 2022: A Delphi poll on its future. (2013, background): “The “prediction paradox” holds that the more humility we have about our ability to make predictions, the more successful we can be in planning for the future (Silver, 2012).”