https://artificialideas.org/2026/11/07/aftershocks-of-an-outage-testing-seismologys-omori-law-against-how-cloud-incidents-cluster/Artificial Ideasen2026-11-07T16:00:00+00:00Aftershocks of an Outage: Testing Seismology’s Omori Law Against How Cloud Incidents Clustercloud outages, site reliability, cascading failures, Omori law, AWS outage, Hawkes process, ETAS, aftershocks, incident response, self-exciting processeshttps://artificialideas.org/wp-content/uploads/2026/11/img_9788.jpg?w=150https://artificialideas.org/2026/11/06/when-a-beats-b-beats-c-beats-a-testing-elo-style-ratings-against-intransitive-strategies-in-games-and-model-leaderboards/Artificial Ideasen2026-11-07T00:00:00+00:00When A Beats B Beats C Beats A: Testing Elo-Style Ratings Against Intransitive Strategies in Games and Model LeaderboardsGame Theory, Elo rating, AlphaStar, KataGo, Bradley-Terry, intransitivity, Chatbot Arena, LLM leaderboards, rock-paper-scissors, evaluationhttps://artificialideas.org/wp-content/uploads/2026/11/img_9779.jpg?w=150https://artificialideas.org/2026/11/06/who-trusts-the-machine-makers-what-polls-from-washington-to-jakarta-say-about-politics-and-ai-companies/Artificial Ideasen2026-11-06T20:01:00+00:00Who Trusts the Machine Makers? What Polls From Washington to Jakarta Say About Politics and AI CompaniesData Centers, AI politics, public opinion, partisan divide, AI regulation, cue-taking, education and ideology, AI companies, global AI attitudes, technology trusthttps://artificialideas.org/wp-content/uploads/2026/11/img_9811.jpg?w=150https://artificialideas.org/2026/11/06/borrowing-strength-testing-actuarial-credibility-theory-against-in-context-learning/Artificial Ideasen2026-11-06T16:00:00+00:00Borrowing Strength: Testing Actuarial Credibility Theory Against In-Context Learningactuarial science, empirical Bayes, Bayesian inference, ridge regression, shrinkage, Bühlmann credibility, in-context learning, transformers, credibility theory, James-Stein estimatorhttps://artificialideas.org/wp-content/uploads/2026/11/img_9769.jpg?w=150https://artificialideas.org/2026/11/05/step-size-is-a-control-problem-testing-adaptive-step-size-control-from-numerical-analysis-against-learning-rate-schedules/Artificial Ideasen2026-11-06T00:00:00+00:00Step Size Is a Control Problem: Testing Adaptive Step-Size Control From Numerical Analysis Against Learning-Rate SchedulesPI controller, gradient descent, learning rate, step size control, edge of stability, Runge-Kutta, numerical analysis, warmup, sharpness, optimizationhttps://artificialideas.org/wp-content/uploads/2026/11/img_9763.jpg?w=150