#cog #bias ## heuristic bias From the beergame[^1] , following bias can be observerd: ### adjustment bias, anchoring bias intercept ### Representative bias ### Adjacent bias using the observed demand values ## automation bias - [[human-factors-in-healthcare-ai-white-paper.pdf]] defines as "Strategies need to be considered to guard against people relying uncritically on the AI, e.g., the use of explanation and training" which is failure to guard. This can be thought as diffusion movement (reinforcing) is much faster than regulating (balancing) i.e. soul much ahead of body. ## inductive bias - [[inductive bias]] ### Generator Approximation bias ### Approximator Optimization bias ### Discriminator Statistical bias ## illusional bias = compromise with optimal conditional on the payoff averaged over the subjective temporal and spatial prior - To avoid posthumous glory of worse than better dynamics and to enjoy priorhumous appreciation of better than worse - related to [[spandrel/prev_post/how-clever-is-random]]as we are using the benefits of locality i.e. prior knowledge ## confirmation bias - [mark rober's 2023 mit speech confirmation bias](https://youtu.be/0FGlsuTnt_U?t=848) for good human relationship - cross river together ## optimism bias mark rober's commencement speech first advice, [naive optimism](https://youtu.be/0FGlsuTnt_U?t=263) - cross river by one rock at a time, first rock (hobby) and let curiosity and passion guide you initial step [^1]: [[Sterman89_ModManagBeh.pdf]] ## [[inductive bias]]